Thompson & Thompson Genetics in Medicine, 8th Edition

CHAPTER 13. The Treatment of Genetic Disease

The understanding of genetic disease at a molecular level, as presented in Chapters 11 and 12, is the foundation of rational therapy. In the coming decades, increasing annotation of the human genome sequence and the catalogue of human genes, as well as gene, RNA, and protein therapy, will have an enormous impact on the treatment of genetic conditions and other disorders. In this chapter, we review established therapies as well as new strategies for treating genetic disease. Our emphasis will be on therapies that reflect the genetic approach to medicine, and our focus is on single-gene diseases, rather than genetically complex disorders.

The objective of treating genetic disease is to eliminate or ameliorate the effects of the disorder, not only on the patient but also on his or her family. The importance of educating the patient is paramount—not only to achieve understanding of the disease and its treatment, but also to ensure compliance with therapy that may be inconvenient and lifelong. The family must be informed about the risk that the disease may occur in other members. Thus genetic counseling is a major component of the management of hereditary disorders and will be dealt with separately, in Chapter 16.

For single-gene disorders due to loss-of-function mutations, treatment is directed to replacing the defective protein, improving its function, or minimizing the consequences of its deficiency. Replacement of the defective gene product (RNA or protein) may be achieved by direct administration, cell or organ transplantation, or gene therapy. In principle, gene therapy or gene editing will be the preferred mode of treatment of some and perhaps many single-gene diseases, once these approaches become routinely safe and effective. However, even when copies of a normal gene can be transferred into the patient to effect permanent cure, the family will need ongoing genetic counseling, carrier testing, and prenatal diagnosis, in many cases for several generations.

Recent discoveries promise many more exciting and dramatic therapies for genetic disease. These achievements include the first cures of inherited disorders using gene therapy, the development of novel small molecule therapies that can restore activity to mutant proteins, and the ability to prevent the clinical manifestations of previously lethal disorders, including lysosomal storage diseases, by protein replacement therapy.

The Current State of Treatment of Genetic Disease

Genetic disease can be treated at any level from the mutant gene to the clinical phenotype (Fig. 13-1). Treatment at the level of the clinical phenotype includes all the medical or surgical interventions that are not unique to the management of genetic disease. Throughout this chapter, we describe the rationale for treatment at each of these levels. For diseases in which the biochemical or genetic defect is known, the approximate frequency with which the most common strategies are employed is shown in Figure 13-2. The current treatments are not necessarily mutually exclusive, although only gene therapy, gene editing, or cell transplantation can potentially provide cures.

image

FIGURE 13-1 The various levels of treatment that are relevant to genetic disease, with the corresponding strategies used at each level. For each level, a disease discussed in the book is given as an example. All the therapies listed are used clinically in many centers, unless indicated otherwise. Hb F, Fetal hemoglobin; mRNA, messenger RNA; PKU, phenylketonuria; RNAi, RNA interference; SCID, severe combined immunodeficiency. SeeSources & Acknowledgments.

image

FIGURE 13-2 Treatment modalities for inborn errors of metabolism. This figure represents the findings of an analysis of the treatment efficacy of 57 inborn errors of metabolism. The total of the nine different approaches used exceeds 100% because more than one treatment can sometimes be used for a given condition. SeeSources & Acknowledgments.

Although powerful advances are being made, the overall treatment of single-gene diseases is presently deficient. A 25-year longitudinal survey of the effectiveness of treatment of 57 inborn errors of metabolism, reflecting the state of the field up to 2008, is shown in Figure 13-3. Note, however, that inborn errors are a group of diseases for which treatment is advanced, in general, compared to most other types of genetic disorders such as those due, for example, to chromosomal abnormalities, imprinting defects, or copy number variation. An encouraging trend over past decades is that treatment is more likely to be successful if the basic biochemical defect is known. In one study, for example, although treatment increased life span in only 15% of all single-gene diseases studied, life span was improved by approximately 50% in the subset of 57 inborn errors in which the cause was known; significant improvements were also observed for other phenotypes, including growth, intelligence, and social adaptation. Thus research to elucidate the genetic and biochemical bases of hereditary disease has a major impact on the clinical outcome.

image

FIGURE 13-3 The effect of treatment of 57 genetic diseases in which the affected gene or biochemical function is known and for which sufficient information was available for analysis in 2008. A quantitative phenotype scoring system was used to evaluate the efficacy of the therapies. The fraction of treatable diseases will have increased to a small extent since this 2008 survey because of the increasing success of enzyme replacement and a few other treatments, including gene therapy. SeeSources & Acknowledgments.

The improving but still unsatisfactory state of treatment of monogenic diseases is due to numerous factors, including the following:

• Gene not identified or pathogenesis not understood. Although more than 3000 genes have been associated with monogenic diseases, the affected gene is still unknown in more than half of these disorders. This fraction will decrease dramatically over the next decade because of the impact of whole-genome and whole-exome sequencing. However, even when the mutant gene in known, knowledge of the pathophysiological mechanism is often inadequate and can lag well behind gene discovery. In phenylketonuria (PKU), for example, despite decades of study, the mechanisms by which the elevation in phenylalanine impairs brain development and function are still poorly understood (see Chapter 12).

• Prediagnostic fetal damage. Some mutations act early in development or cause irreversible pathological changes before they are diagnosed. These problems can sometimes be anticipated if there is a family history of the genetic disease or if carrier screening identifies couples at risk. In some cases, prenatal treatment is possible (Table 13-1).

TABLE 13-1

Examples of Prenatal Medical Treatment of Monogenic Disorders

Disease

Treatment

Biotinidase deficiency

Prenatal biotin administration

Cobalamin-responsive methylmalonic aciduria

Prenatal maternal cobalamin administration

Congenital adrenal hyperplasia

Dexamethasone, a cortisol analogue

Phosphoglycerate dehydrogenase (PGDH) deficiency, a disorder of L-serine synthesis

Prenatal L-serine administration

• Severe phenotypes are less amenable to intervention. The initial cases of a disease to be recognized are usually the most severely affected, but they are often less amenable to treatment. In such individuals, the mutation frequently leads to the absence of the encoded protein or to a severely compromised mutant protein with no residual activity. In contrast, when the mutation is less disruptive, the mutant protein may retain some residual function and it may be possible to increase the small amount of function sufficiently to have a therapeutic effect, as described later.

• The challenge of dominant negative alleles. For some dominant disorders, the mutant protein interferes with the function of the normal allele. The challenge is to decrease the expression or impact of the mutant allele or its encoded mutant protein specifically, without disrupting expression or function of the normal allele or its normal protein.

Special Considerations in Treating Genetic Disease

Long-Term Assessment of Treatment Is Critical

For treating monogenetic diseases, long-term evaluation of cohorts of treated individuals, often over decades, is critical for several reasons. First, treatment initially judged as successful may eventually be revealed to be imperfect; for example, although well-managed children with PKU have escaped severe retardation and have normal or nearly normal IQs (see later), they often manifest subtle learning disorders and behavioral disturbances that impair their academic performance in later years.

Second, successful treatment of the pathological changes in one organ may be followed by unexpected problems in tissues not previously observed to be clinically involved, because the patients typically did not survive long enough for the new phenotype to become evident. Galactosemia, a well-known inborn error of carbohydrate metabolism, illustrates this point. This disorder results from an inability to metabolize galactose, a component of lactose (milk sugar), because of the autosomal recessive deficiency of galactose-1-phosphate uridyltransferase (GALT)

image

Affected infants are usually normal at birth but develop gastrointestinal problems, cirrhosis of the liver, and cataracts in the weeks after they are given milk. The pathogenesis is thought to be due to the negative impact of galactose-1-phosphate accumulation on other critical enzymes. If not recognized, galactosemia causes severe intellectual disability and is often fatal. Complete removal of milk from the diet, however, can protect against most of the harmful consequences, although, as with PKU, learning disabilities are now recognized to be common, even in well-treated patients. Moreover, despite conscientious treatment, most females with galactosemia have ovarian failure that appears to result from continued galactose toxicity.

Another example is provided by hereditary retinoblastoma (Case 39) due to germline mutations in the retinoblastoma (RB1) gene (see Chapter 15). Patients successfully treated for the eye tumor in the first years of life are unfortunately at increased risk for development of other independent malignant neoplasms, particularly osteosarcoma, after the first decade of life. Ironically, therefore, treatment that successfully prolongs life provides an opportunity for the manifestation of a previously unrecognized phenotype.

In addition, therapy that is free of side effects in the short term may be associated with serious problems in the long term. For example, clotting factor infusion in hemophilia (Case 21) sometimes results in the formation of antibodies to the infused protein, and blood transfusion in thalassemia (Case 44) invariably produces iron overload, which must then be managed by the administration of iron-chelating agents, such as deferoxamine.

Genetic Heterogeneity and Treatment

The optimal treatment of single-gene defects requires an unusual degree of diagnostic precision; one must often define not only the biochemical abnormality, but also the specific gene that is affected. For example, as we saw in Chapter 12, hyperphenylalaninemia can result from mutations in either the phenylalanine hydroxylase (PAH) gene or in one of the genes that encodes the enzymes required for the synthesis of tetrahydrobiopterin (BH4), the cofactor of the PAH enzyme (see Fig. 12-2). The treatment of these two different causes of hyperphenylalaninemia is entirely different, as shown previously in Table 12-1.

Allelic heterogeneity (see Chapter 7) may also have critical implications for therapy. Some alleles may produce a protein that is decreased in abundance but has some residual function, so that strategies to increase the expression, function, or stability of such a partially functional mutant protein may correct the biochemical defect. This situation is again illustrated by some patients with hyperphenylalaninemia due to mutations in the PAH gene; the mutations in some patients lead to the formation of a mutant PAH enzyme whose activity can be increased by the administration of high doses of the BH4 cofactor (see Chapter 12). Of course, if a patient carries two alleles with no residual function, nothing will be gained by increasing the abundance of the mutant protein. One of the most striking examples of the importance of knowing the specific mutant allele in a patient with a genetic disease is exemplified by cystic fibrosis (CF); the drug ivacaftor (Kalydeco) is presently approved for treating CF patients carrying any one of only nine of the many hundreds of CFTR missense alleles.

Treatment by the Manipulation of Metabolism

Presently, the most successful disease-specific approach to the treatment of genetic disease is directed at the metabolic abnormality in inborn errors of metabolism. The principal strategies used to manipulate metabolism in the treatment of this group of diseases are listed in Table 13-2. The necessity for patients with pharmacogenetic diseases, such as glucose-6-phosphate dehydrogenase deficiency, to avoid certain drugs and chemicals is described in Chapter 18.

TABLE 13-2

Treatment of Genetic Disease by Metabolic Manipulation

Type of Metabolic Intervention

Substance or Technique

Disease

Avoidance

Antimalarial drugs

G6PD deficiency

Isoniazid

Slow acetylators

Dietary restriction

Phenylalanine

PKU

Galactose

Galactosemia

Replacement

Thyroxine

Biotin

Monogenic forms of congenital hypothyroidism

Biotinidase deficiency

Diversion

Sodium benzoate

Urea cycle disorders

Drugs that sequester bile acids in the intestine (e.g., colesevelam)

Familial hypercholesterolemia heterozygotes

Enzyme inhibition

Statins

Familial hypercholesterolemia heterozygotes

Receptor antagonism

Losartan (investigational)

Marfan syndrome

Depletion

LDL apheresis (direct removal of LDL from plasma)

Familial hypercholesterolemia homozygotes

G6PD, Glucose-6-phosphate dehydrogenase; LDL, low-density lipoprotein; PKU, phenylketonuria.

Updated from Rosenberg LE: Treating genetic diseases: lessons from three children. Pediatr Res 27:S10–S16, 1990.

Substrate Reduction

As illustrated by the damaging effects of hyperphenylalaninemia in PKU, enzyme deficiencies may lead to substrate accumulation, with pathophysiological consequences (see Chapter 12). Strategies to prevent the accumulation of the offending substrate have been one of the most effective methods of treating genetic disease. The most common approach is to reduce the dietary intake of the substrate or of a precursor of it, and presently several dozen disorders—most involving amino acid catabolic pathways—are managed in this way. The drawback is that severe lifelong restriction of dietary protein intake is often necessary, requiring strict adherence to an artificial diet that is onerous for the family as well as for the patient. Nutrients such as 20 essential amino acids cannot be withheld entirely, however; their intake must be sufficient for anabolic needs such as protein synthesis.

A diet restricted in phenylalanine largely circumvents the neurological damage in classic PKU (see Chapter 12). Phenylketonuric children are normal at birth because the maternal enzyme protects them during prenatal life. Treatment is most effective if begun promptly after diagnosis by newborn screening. Without treatment, irreversible developmental delay occurs, the degree of intellectual deficit being directly related to the delay in commencing the low-phenylalanine diet. It is now recommended that patients with PKU remain on a low-phenylalanine diet for life because neurological and behavioral abnormalities develop in many (although perhaps not all) patients if the diet is stopped. However, even PKU patients who have been effectively treated throughout life may have neuropsychological deficits (e.g., impaired conceptual, visual-spatial, and language skills), despite their having normal intelligence as measured by IQ tests. Nonetheless, treatment produces results vastly superior to the severe developmental delay that occurs without treatment. As discussed in Chapter 12, continued phenylalanine restriction is particularly important in women with PKU during pregnancy to prevent prenatal damage to the fetus, even though the fetus is highly unlikely to be affected by PKU.

Replacement

The provision of essential metabolites, cofactors, or hormones whose deficiency is due to a genetic disease is simple in concept and often simple in application. Some of the most successfully treated single-gene defects belong to this category. A prime example is provided by congenital hypothyroidism, of which 10% to 15% of cases are monogenic in origin. Monogenic congenital hypothyroidism can result from mutations in any one of numerous genes encoding proteins required for the development of the thyroid gland or the biosynthesis or metabolism of thyroxine. Because congenital hypothyroidism from all causes is common (approximately 1 in 4000 neonates), neonatal screening is conducted in many countries so that thyroxine administration may be initiated soon after birth to prevent the severe intellectual defects that are otherwise inevitable (see Chapter 18).

Diversion

Diversion therapy is the enhanced use of alternative metabolic pathways to reduce the concentration of a harmful metabolite. A major use of this strategy is in the treatment of the urea cycle disorders (Fig. 13-4). The function of the urea cycle is to convert ammonia, which is neurotoxic, to urea, a benign end product of protein catabolism excreted in urine. If the cycle is disrupted by an enzyme defect such as ornithine transcarbamylase deficiency (Case 36), the consequent hyperammonemia can be only partially controlled by dietary protein restriction. Blood ammonia levels can be reduced to normal, however, by the diversion of ammonia to metabolic pathways that are normally of minor significance, leading to the synthesis of harmless compounds. Thus, the administration to hyperammonemic patients of large quantities of sodium benzoate forces the ligation of ammonia with glycine to form hippurate, which is excreted in urine (see Fig. 13-4). Glycine synthesis is thereby increased, and for each mole of glycine formed, one mole of ammonia is consumed.

image

FIGURE 13-4 The strategy of metabolite diversion. In this example, ammonia cannot be removed by the urea cycle because of a genetic defect of a urea cycle enzyme. The administration of sodium benzoate diverts ammonia to glycine synthesis, and the nitrogen moiety is subsequently excreted as hippurate.

A comparable approach is used to reduce cholesterol levels in heterozygotes for familial hypercholesterolemia (Case 16) (see Chapter 12). If bile acids are sequestered in the intestine by the oral administration of a compound such as colesevelam and then excreted in feces rather than being reabsorbed, bile acid synthesis from cholesterol increases (Fig. 13-5). The reduction in hepatic cholesterol levels leads to increased production of low-density lipoprotein (LDL) receptors from their single normal LDL receptor gene, increased hepatic uptake of LDL-bound cholesterol, and lower levels of plasma LDL cholesterol. This treatment significantly reduces plasma cholesterol levels because 70% of all LDL receptor uptake of cholesterol occurs in the liver. An important general principle is illustrated by this example: autosomal dominant diseases may sometimes be treated by increasing the expression of the normal allele.

image

FIGURE 13-5 Rationale for the combined use of a reagent that sequesters bile acids, such as colesevelam, together with an inhibitor of 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMG CoA reductase) in the treatment of familial hypercholesterolemia heterozygotes. LDL, Low-density lipoprotein. SeeSources & Acknowledgments.

Enzyme Inhibition

The pharmacological inhibition of enzymes is sometimes used to reduce the impact of metabolic abnormalities in treating inborn errors. This principle is also illustrated by the treatment of heterozygotes of familial hypercholesterolemia. If a statin, a class of drugs that are powerful inhibitors of 3-hydroxy-3-methylglutaryl coenzyme A reductase, or HMG CoA reductase (the rate-limiting enzyme of cholesterol synthesis), is used to decrease hepatic de novo cholesterol synthesis in these patients, the liver compensates by increasing the synthesis of LDL receptors from the remaining intact LDL receptor allele. The increase in LDL receptors typically lowers plasma LDL cholesterol levels by 40% to 60% in familial hypercholesterolemia heterozygotes; used together with colesevelam, the effect is synergistic, and even greater decreases can be achieved (see Fig. 13-5).

Receptor Antagonism

In some instances, the pathophysiology of an inherited disease results from the increased and inappropriate activation of a biochemical or signaling pathway. In such cases, one therapeutic approach is to antagonize critical steps in the pathway. A powerful example is provided by an investigational treatment of an autosomal dominant connective tissue disorder, Marfan syndrome (Case 30). The disease results from mutations in the gene that encodes fibrillin 1, an important structural component of the extracellular matrix. The syndrome is characterized by many connective tissue abnormalities, such as aortic aneurysm, pulmonary emphysema, and eye-lens dislocation (Fig. 13-6).

image

FIGURE 13-6 Magnetic resonance image (MRI) of the abdominal aorta of a 29-year-old pregnant woman with Marfan syndrome. The massive dilatation of the abdominal aorta is indicated by the arrowSeeSources & Acknowledgments.

Unexpectedly, the pathophysiology of Marfan syndrome is only partially explained by the impact of the reduction in fibrillin-1 microfibrils on the structure of the extracellular matrix. Rather, it has been found that a major function of microfibrils is to regulate signaling by the transforming growth factor β (TGF-β), by binding TGF-β to the large latent protein complex of TGF-β. The decreased abundance of microfibrils in Marfan syndrome leads to an increase in the local abundance of unbound TGF-β and in local activation of TGF-β signaling. This increased TGF-β signaling has been suggested to underlie the pathogenesis of many of the phenotypes of Marfan syndrome, particularly the progressive dilation of the aortic root, and aortic aneurysm and dissection, the major cause of death in this disorder. Moreover, a recently recognized group of other vasculopathies, such as nonsyndromic forms of thoracic aortic aneurysm, has also proved to be driven by altered TGF-β signaling.

Angiotensin II signaling is known to increase TGF-β activity and the angiotensin II type 1 receptor antagonist, losartan, a widely used antihypertensive agent, has been shown to attenuate TGF-β signaling by decreasing the transcription of genes encoding TGF-β ligands, receptor subunits, and activators. Treatment with losartan has been found to decrease substantially the rate of aortic root dilation in initial clinical trials of Marfan syndrome patients, an effect that appears to be largely due to decreased TGF-β signaling.

The novel use of a U.S. Food and Drug Administration (FDA) approved drug, losartan, to treat a rare inherited disease, Marfan syndrome, is likely to represent a paradigm that will be repeated regularly in the future, as small molecule chemical screens to identify compounds with therapeutic potential—often including the thousands of FDA approved drugs—are undertaken to identify safe, effective treatments for other uncommon genetic disorders.

Depletion

Genetic diseases characterized by the accumulation of a harmful compound are sometimes treated by direct removal of the compound from the body. This principle is illustrated by the treatment of homozygousfamilial hypercholesterolemia. In this instance, for patients whose LDL levels cannot be lowered by other approaches, a procedure called apheresis is used to remove LDL from the circulation. Whole blood is removed from the patient, LDL is removed from plasma by any one of several methods, and the plasma and blood cells are returned to the patient. The use of phlebotomy to alleviate the iron accumulation of hereditary hemochromatosis (Case 20) provides another example of depletion therapy.

Treatment to Increase the Function of the Affected Gene or Protein

The growth in knowledge of the molecular pathophysiology of monogenic diseases has been accompanied by a small but promising increase in therapies that—at the level of DNA, RNA, or protein—increase the function of the gene affected by the mutation. Some of the novel treatments have led to striking improvement in the lives of affected individuals, an outcome that, until recently, would have seemed fanciful. An overview of the molecular treatment of single-gene diseases is presented in Figure 13-7. These molecular therapies represent one facet of the important paradigm embraced by the concept of personalized or precision medicine. The term precision medicine is a general one used to describe the diagnosis, prevention, and treatment of a disease—tailored to individual patients—based on a profound understanding of the mechanisms that underlie its etiology and pathogenesis.

image

FIGURE 13-7 The molecular treatment of inherited disease. Each molecular therapy is discussed in the text. ADA, Adenosine deaminase; ASO, antisense oligonucleotide; ERT, enzyme replacement therapy; Hb F, fetal hemoglobin; mRNA, messenger RNA; MSD, membrane-spanning domain; NBD, nucleotide-binding domain; PEG, polyethylene glycol; SCID, severe combined immunodeficiency; siRNA, small interfering RNA.

Treatment at the Level of the Protein

In many situations, if a mutant protein product is made, it may be possible to increase its function. For example, the stability or function of a mutant protein with some residual function may be further increased. With enzymopathies, the improvement in function obtained by this approach is usually very small, on the order of a few percent, but this increment is often all that is required to restore biochemical homeostasis.

Enhancement of Mutant Protein Function with Small Molecule Therapy

Small molecules are compounds with molecular weights in the few hundreds to thousands. They include vitamins, nonpeptide hormones, and indeed most drugs, whether synthesized by organic chemists or isolated from nature. A new strategy for identifying potential drugs is to use high-throughput screening of chemical compound libraries, often containing tens of thousands of known chemicals, against a drug target, such as the protein whose function is disrupted by a mutation. As we will discuss, two drugs that are now FDA approved for the treatment of some patients with CF, and another that is investigational, were discovered using such high-throughput screens. Progress in the development of these drugs represents a new frontier with great potential for the treatment of genetic disease.

Small Molecule Therapy to Allow Skipping over Nonsense Codons.

Nonsense mutations account for 11% of defects in the human genome. Approximately 9% of all CFTR alleles are nonsense mutations, and approximately 50% of Ashkenazi Jewish patients with CF carry at least one CFTR allele with a premature stop codon (e.g., Arg553Stop). A potentially ideal therapeutic approach (other than gene therapy) for patients with a nonsense mutation would be a safe drug that encourages the translational apparatus to misread the stop codon by a transfer RNA (tRNA) that is near-cognate to the stop codon tRNA. If the amino acid thereby inserted into the polypeptide by that tRNA still produces a functional protein, the activity of the protein would be restored. An event of this type, for example, would convert the CFTR Arg553Stop mutation to 553Tyr, a substitution that generates a CFTR peptide with nearly normal properties. High-throughput chemical screens for a drug of this type identified ataluren (PTC124), and evidence suggests that it is most effective in allowing read-through of TGA nonsense codons. Moreover, studies in model organisms have firmly demonstrated that it can correct the mutant phenotype of some nonsense mutations. Ataluren has not been established to be clinically effective, but a Phase III clinical trial in CF patients carrying at least one nonsense mutation showed a promising trend toward statistically significant improvement in lung function, and a follow-up trial is underway. Even if ataluren proves ineffective in humans, thousands of other small molecules are being examined in laboratories around the world to identify novel nontoxic compounds that facilitate the skipping of nonsense codons, not only for the treatment of CF but also for Duchenne muscular dystrophy patients carrying nonsense codons, as well as other diseases. Safe, effective drugs of this type will have a major impact on the treatment of inherited disease.

Small Molecules to Correct the Folding of Mutant Membrane Proteins: Pharmacological Chaperones.

Some mutations in membrane proteins may disrupt their ability to fold, pass through the endoplasmic reticulum, and be trafficked to the plasma membrane. These mutant proteins are recognized by the cellular protein quality control machinery, trapped in the endoplasmic reticulum, and prematurely degraded by the proteosome. The ΔF508 deletion of the CFTR protein—which constitutes 65% of all CF mutations worldwide—is perhaps the best-known example (see Fig. 12-15) of a mutation that impairs trafficking of a membrane protein. If the folding/trafficking defect could be overcome to increase the abundance of CFTR channels at the apical surface of the cell by 20% to 25%, it is thought that a clinical benefit would be obtained, because once the ΔF508 CFTR protein reaches the cell surface, it is an effective Cl channel.

Small molecule screens to identify compounds that can serve as a chaperone to prevent misfolding and correct the ΔF508 CFTR trafficking defect in in vitro assay systems have identified lumacaftor (VX-809) as an effective, although incomplete, corrector of this specific CFTR mutant polypeptide (see Fig. 13-7). Lumacaftor interacts directly with the mutant CFTR to stabilize its three-dimensional structure, specifically correcting the underlying trafficking defect and enhancing Cl transport. Although monotherapy with lumacaftor had no clinical benefits, a recently completed Phase III clinical trial using lumacaftor together with another small molecule, ivacaftor (VX-770), discussed later, showed significant improvements in lung function in homozygous ΔF508 CFTR patients. This finding is notable because it is the first treatment shown to have a favorable impact on the primary biochemical defect in patients carrying the most common CFTR allele, ΔF508. Ongoing studies of the long-term effectiveness and safety of the lumacaftor-ivacaftor combination therapy are in progress. Irrespective of their success, this example is a milestone in medical genetics, because it establishes the principle that molecular chaperones can have clinical benefits in the treatment of monogenic disease.

Small Molecules to Increase the Function of Correctly Trafficked Mutant Membrane Proteins.

Amino acid substitutions in membrane proteins may not disrupt the trafficking of the mutant polypeptide to the plasma membrane, but rather interfere with its function at the cell surface. Small molecule screens for new treatments for CF have also led this area of drug discovery. Screens for so-called potentiators—molecules that could enhance the function of mutant CFTR proteins that are correctly positioned at the cell surface—identified ivacaftor (VX-770), which improves the Cl transport of some mutant CFTR proteins, such as the Gly551Asp CFTR missense mutation (see Fig. 12-15) that inactivates anion transport; this allele is carried by 4% to 5% of all CF patients. In one clinical trial, patients carrying at least one Gly551Asp allele experienced a significant improvement in lung function (Fig. 13-8), weight gain, respiratory symptoms, and a decline in sweat Cl. Ivacaftor is presently FDA approved for the treatment of eight other CFTR missense mutations, and more alleles will certainly be added to this group. Although fewer than 200 CF patients in the United States have one of these eight alleles, the allele-specific indications for ivacaftor treatment highlight both the benefits and dilemmas of personalized medicine for genetic disease: effective drugs can be discovered, but they may be effective only in a relatively small numbers of individuals. Moreover, at present ivacaftor is extremely expensive, costing approximately $300,000 per year.

image

FIGURE 13-8 The effect of ivacaftor (Kalydeco) on lung function of cystic fibrosis patients carrying at least one Gly551Asp CFTR allele. The figure shows the absolute mean change from baseline in the percent of predicted forced expiratory volume in 1 second (FEV1) through week 48 of a clinical trial. N refers to the number of subjects studied at each time point during the trial. SeeSources & Acknowledgments.

Small Molecules to Enhance the Function of Mutant Enzymes: Vitamin-Responsive Inborn Errors of Metabolism.

The biochemical abnormalities of a number of inherited metabolic diseases may respond, sometimes dramatically, to the administration of large amounts of the vitamin cofactor of the enzyme impaired by the mutation (Table 13-3). In fact, the vitamin-responsive inborn errors are among the most successfully treated of all genetic diseases. The vitamins used are remarkably nontoxic, generally allowing the safe administration of amounts 100 to 500 times greater than those required for normal nutrition. In homocystinuria due to cystathionine synthase deficiency (see Fig. 12-8), for example, approximately 50% of patients respond to the administration of high doses of pyridoxine (vitamin B6, the precursor of pyridoxal phosphate, the cofactor for the enzyme), an example—as we saw earlier in the case of BH4administration in PKU—of cofactor responsiveness in a metabolic disease. In most of these responsive patients, homocystine completely disappears from the plasma, even though the increase in hepatic cystathionine synthase activity is usually only a fewfold, from 1.5% to 4.5% of control activity. The increased pyridoxal phosphate concentrations may stabilize the mutant enzyme or overcome reduced affinity of the mutant enzyme for the cofactor (Fig. 13-9). In any case, vitamin B6 treatment substantially improves the clinical course of the disease in responsive patients. Nonresponsive patients generally carry null alleles and therefore have no residual cystathionine synthase activity to augment.

TABLE 13-3

Treatment of Genetic Disease at the Level of the Mutant Protein

Strategy

Example

Status

Enhancement of Mutant Protein Function

Small molecules that facilitate translational “skipping” over mutant stop codons

Ataluren in the 10% of cystic fibrosis patients with nonsense mutations in the CFTR gene

Investigational in CF: confirmatory Phase III clinical trial was begun in 2014

Small molecule “correctors” that increase the trafficking of the mutant protein through the ER to the plasma membrane

Lumacaftor (VX-809) to increase the abundance of the ΔF508 mutant CFTR protein at the apical membrane of epithelial cells in CF patients

Investigational: very promising improvements in lung function in ΔF508 homozygotes, when used in combination with ivacaftor; expensive

Small molecule “potentiators” that increase the function at the cell membrane of correctly trafficked membrane proteins

Ivacaftor (VX-770) used alone to enhance the function of specific mutant CFTR proteins at the epithelial apical membrane

FDA approved for the treatment of CF patients carrying specific alleles; expensive

Vitamin cofactor administration to increase the residual activity of the mutant enzyme

Vitamin B6 for pyridoxine-responsive homocystinuria

Treatment of choice in the 50% of cystathionine synthase patients who are responsive

Protein Augmentation

Replacement of an extracellular protein

Factor VIII in hemophilia A

Well-established, effective, safe

Extracellular replacement of an intracellular protein

Polyethylene glycol–modified adenosine deaminase (PEG-ADA) in ADA deficiency

Well-established, safe, and effective, but costly; now used principally to stabilize patients before gene therapy or HLA-matched bone marrow transplantation

Replacement of an intracellular protein—cell targeting

β-glucocerebrosidase in non-neuronal Gaucher disease

Established; biochemically and clinically effective; expensive

ADA, Adenosine deaminase; CF, cystic fibrosis; ER, endoplasmic reticulum; FDA, U.S. Food and Drug Administration; HLA, human leukocyte antigen; PEG, polyethylene glycol.

image

FIGURE 13-9 The mechanism of response of a mutant apoenzyme to the administration of its cofactor at high doses. Vitamin-responsive enzyme defects are often due to mutations that reduce the normal affinity (top) of the enzyme protein (apoenzyme) for the cofactor needed to activate it. In the presence of the high concentrations of the cofactor that result from the administration of up to 500 times the normal daily requirement, the mutant enzyme acquires a small amount of activity sufficient to restore biochemical normalcy. SeeSources & Acknowledgments.

Protein Augmentation

The principal types of protein augmentation are summarized in Table 13-3. Protein augmentation is a routine therapeutic approach in only a few diseases, all involving proteins whose principal site of action is in the plasma or extracellular fluid. The prime example is the prevention or arrest of bleeding episodes in patients with hemophilia (Case 21) by the infusion of plasma fractions enriched for the appropriate factor. The decades of experience with this disease illustrate the problems that can be anticipated as new strategies for replacing other, particularly intracellular, polypeptides are attempted. These problems include the difficulty and cost of procuring sufficient amounts of the protein to treat all patients at the optimal frequency, the need to administer the protein at a frequency consistent with its half-life (only 8 to 10 hours for factor VIII), and the formation of neutralizing antibodies in some patients (5% of classic hemophiliacs).

Enzyme Replacement Therapy: Extracellular Administration of an Intracellular Enzyme

Adenosine Deaminase Deficiency.

Adenosine deaminase (ADA) is a critical enzyme of purine metabolism that catalyzes the deamination of adenosine to inosine and of deoxyadenosine to deoxyinosine (Fig. 13-10). The pathology of ADA deficiency, an autosomal recessive disease, results entirely from the accumulation of toxic purines, particularly deoxyadenosine, in lymphocytes. A profound failure of both cell-mediated (T-cell) and humoral (B-cell) immunity results, making ADA deficiency one cause of severe combined immunodeficiency (SCID). Untreated patients die of infection within the first 2 years of life. The long-term treatment of ADA deficiency is rapidly evolving, with gene therapy (see later section) now a strong alternative to bone marrow transplantation from a fully human leukocyte antigen (HLA) compatible donor. The administration of a modified form of the bovine ADA enzyme, described in the next section, is no longer a first choice for long-term management, but it is an effective stabilizing measure in the short term until these other treatments can be used.

image

FIGURE 13-10 Adenosine deaminase (ADA) converts adenosine to inosine and deoxyadenosine to deoxyinosine. In ADA deficiency, deoxyadenosine accumulation in lymphocytes is lymphotoxic, killing the cells by impairing DNA replication and cell division to cause severe combined immunodeficiency (SCID).

Modified Adenosine Deaminase.

The infusion of bovine ADA modified by the covalent attachment of an inert polymer, polyethylene glycol (PEG), is superior in several ways to the use of the unmodified ADA enzyme. First, PEG-ADA largely protects the patient from a neutralizing antibody response (which would remove the ADA from plasma). Second, the modified enzyme remains in the extracellular fluid where it can degrade toxic purines. Third, the plasma half-life of PEG-ADA is 3 to 6 days, much longer than the half-life of unmodified ADA. Although the near-normalization of purine metabolism obtained with PEG-ADA does not completely correct immune function (most patients remain T lymphopenic), immunoprotection is restored, with dramatic clinical improvement.

The general principles exemplified by the use of PEG-ADA are that (1) proteins can be chemically modified to improve their effectiveness as pharmacological reagents, and (2) an enzyme that is normally located inside the cell can be effective extracellularly if its substrate is in equilibrium with the extracellular fluid and if its product can be taken up by the cells that require it.

Enzyme Replacement Therapy: Targeted Augmentation of an Intracellular Enzyme.

Enzyme replacement therapy (ERT) is now established therapy for six lysosomal storage diseases, with clinical trials being conducted for several others. Non-neuronal (type 1) Gaucher disease was the first lysosomal storage disease for which ERT was shown to be effective. It is the most prevalent lysosomal storage disorder, affecting up to 1 in 450 Ashkenazi Jews and 1 in 40,000 to 100,000 individuals in other populations (Case 18). This autosomal recessive condition results from deficiency of β-glucocerebrosidase. Loss of this enzyme activity leads to the accumulation of its substrate, the complex lipid glucocerebroside, in the lysosome, where it is normally degraded. The lysosomal accumulation of glucocerebroside, particularly in the macrophages and monocytes of the reticuloendothelial system, leads to gross enlargement of the liver and spleen. Bone marrow is slowly replaced by lipid-laden macrophages (Gaucher cells), leading to anemia and thrombocytopenia. The bone lesions cause episodic pain, osteonecrosis, and substantial morbidity.

More than 5000 patients with non-neuronal Gaucher disease have been treated worldwide with β-glucocerebrosidase ERT, with dramatic clinical benefits. The increase in the hemoglobin level of one patient, a response that is representative of the effectiveness of this treatment, is shown in Figure 13-11. Overall, this therapy also reduces the enlargement of liver and spleen, increases the platelet count, accelerates growth, and improves the characteristic skeletal abnormalities and bone density. Early treatment is most effective in preventing irreversible damage to bones and liver.

image

FIGURE 13-11 The effect of weekly intravenous infusions of modified glucocerebrosidase on the hemoglobin concentration of a child with non-neuronal (type 1) Gaucher disease. A review of the response of more than 1000 patients indicates that this response is representative. Treatment was begun at 4 years of age and continued for 18 months. The therapy was accompanied by an increased platelet count and radiological improvement in the bone abnormalities. The hematological parameters returned to pretreatment levels when the infusions were stopped. SeeSources & Acknowledgments.

The success of ERT for non-neuronopathic Gaucher disease provides guidance in the development of enzyme and protein replacement therapy for other lysosomal storage disorders, and perhaps other classes of diseases as well, for several reasons. First, this use of ERT highlights the importance of understanding the biology of the relevant cell types. As demonstrated by I-cell disease (see Chapter 12), lysosomal hydrolases such as β-glucocerebrosidase contain post-translationally added mannose sugars that target the enzyme to the macrophage through a mannose receptor on the plasma membrane. Once bound, the enzyme is internalized and delivered to the lysosome. Thus, β-glucocerebrosidase ERT in Gaucher disease targets the protein both to a particular relevant cell and to a specific intracellular address, in this case the macrophage and the lysosome, respectively.

Second, the human enzyme can be produced in abundance from cultured cells expressing the glucocerebrosidase gene, a key factor because this treatment, given as twice-monthly infusions, must be continuous. Only approximately 1% to 5% of the normal intracellular enzyme activity is required to correct the biochemical abnormalities in this and other lysosomal storage disorders. Third, the administered β-glucocerebrosidase is not recognized as a foreign antigen because patients with non-neuronal Gaucher disease have small amounts of residual enzyme activity. Unfortunately, however, because β-glucocerebrosidase does not cross the blood-brain barrier, ERT cannot treat the neuronopathic forms of Gaucher disease. Although ERT for any lysosomal disease is very expensive, its success has been a tremendous advance in the treatment of monogenic disorders. It has established the feasibility of directing an intracellular enzyme to its physiologically relevant location to produce clinically significant effects.

Modulation of Gene Expression

Decades ago, the idea that one might treat a genetic disease through the use of drugs that modulate gene expression would have seemed fanciful. Increasing knowledge of the normal and pathological bases of gene expression, however, has made this approach feasible. Indeed, it seems likely that this strategy will become only more widely used as our understanding of gene expression, and how it might be manipulated, increases.

Increasing Gene Expression from the Wild-Type or Mutant Locus

Therapeutic effects can be obtained by increasing the amount of messenger RNA (mRNA) transcribed from the wild-type locus associated with a dominant disease or from the mutant locus, if the mutant protein retains some function (Table 13-4; see Fig. 13-7). An effective therapy of this type is used to manage hereditary angioedema, a rare but potentially fatal autosomal dominant condition due to mutations in the gene encoding the complement 1 (C1) esterase inhibitor. Affected individuals are subject to unpredictable episodes, of widely varying severity, of submucosal and subcutaneous edema. Attacks that involve the upper respiratory tract can be fatal. Because of the rapid and unpredictable nature of the attacks, long-term prophylaxis with attenuated androgens, particularly danazol, is often employed. Danazol significantly increases the abundance of the C1 esterase inhibitor mRNA by modulating transcription of the gene, presumably from both the normal and mutant loci. In the great majority of patients, the frequency of serious attacks is dramatically reduced, although long-term androgen administration is not free of side effects.

TABLE 13-4

Treatment by Modification of the Genome or its Expression

image

cas, CRISPR-associated; CRISPR, clustered regularly interspaced short palindromic repeats; Hb F, fetal hemoglobin; HLA, human leukocyte antigen.

Increasing Gene Expression from a Locus Not Affected by the Disease

A related therapeutic strategy is to increase the expression of a normal gene that compensates for the effect of mutation at another locus. This approach is extremely promising in the management of sickle cell disease (Case 42) and β-thalassemia (Case 44), for which drugs that induce DNA hypomethylation are being used to increase the abundance of fetal hemoglobin (Hb F) (see Chapter 11), which normally constitutes less than 1% of total hemoglobin in adults. Sickle cell disease causes illness because of both the anemia and the sickling of red blood cells (see Chapter 11); the increase in the level of Hb F (α2γ2) benefits these patients because Hb F is a perfectly adequate oxygen carrier in postnatal life and because the polymerization of deoxyhemoglobin S is inhibited by Hb F. In β-thalassemia, Hb F restores the imbalance between α and non–α-globin chains (see Chapter 11), substituting Hb F (α2γ2) for Hb A (α2β2).

The normal postnatal decrease in the expression of the γ-globin gene is at least partly due to methylation of CpG residues (see Chapter 3) in the promoter region of the gene. Methylation of the promoter is inhibited if a cytidine analogue such as decitabine (5-aza-2′-deoxycytidine) is incorporated into DNA instead of cytidine. The inhibition of methylation is associated with substantial increases in γ-globin gene expression and, accordingly, in the proportion of Hb F in blood. Both patients with sickle cell anemia and patients with some forms of β-thalassemia treated with decitabine uniformly display increases in Hb F to levels that are likely to have a significant positive impact on morbidity and mortality (Fig. 13-12). The use of inhibitors of γ-globin gene methylation is evolving rapidly, and more effective inhibitors of methylation, with fewer side effects, are likely to be developed.

image

FIGURE 13-12 The effect of the cytosine analogue decitabine, a DNA hypomethylating agent, on the percentage of fetal hemoglobin (Hb F) in 13 patients with sickle cell disease, compared with their level of Hb F without any treatment. Note the wide variation between patients in the levels of Hb F without treatment. Every patient shown had a significant increase in Hb F during decitabine therapy. SeeSources & Acknowledgments.

As described earlier, any approach that allows a patient with β-thalassemia or sickle cell anemia to retain Hb F expression is likely to be very beneficial to the patient. The BCL11A protein, described in Chapter 11, is a trans-acting effector of hemoglobin switching that turns off γ-globin production postnatally but nevertheless allows β-globin gene expression. Genome editing (see later) in hematopoietic stem cells (HSCs) is currently being explored as a method to delete an erythroid enhancer of the BCL11A gene, thereby blocking its expression in the erythroid cell lineage. As a result, hemoglobin switching from Hb F to Hb A would not occur, and patients would retain Hb F instead of a hemoglobin containing a mutant β-thalassemia or sickle cell allele.

Reducing the Expression of a Dominant Mutant Gene Product: Small Interfering RNAs

The pathology of some inherited diseases results from the presence of a mutant protein that is toxic to the cell, as seen with proteins with expanded polyglutamine tracts (see Chapter 12), as in Huntington disease(Case 24), or with disorders such as the inherited amyloidoses. The autosomal dominant disorder transthyretin amyloidosis is the result of any of more than 100 missense mutations in transthyretin, a protein produced mainly in liver, that transports retinol (one form of vitamin A) and thyroxine in body fluids. The major phenotypes are amyloidotic polyneuropathy, due to deposition of the amyloid in peripheral nerves (causing intractable peripheral sensory neuropathy and autonomic neuropathy), and amyloidotic cardiomyopathy, due to its deposition in the heart. Both disorders greatly shorten the life span, and the only current treatment is hepatic transplantation.

A promising therapy, however, is provided by a technology called RNA interference (RNAi), which can mediate the degradation of a specific target RNA, such as that encoding transthyretin. Briefly, short RNAs that correspond to specific sequences of the targeted RNA (see Fig. 13-7)—termed small interfering RNAs (siRNAs)—are introduced into cells by, for example, lipid nanoparticles or viral vectors. Strands of the interfering RNA, approximately 21 nucleotides long, bind to the target RNA and initiate its cleavage. A Phase I clinical trial using an siRNA (encapsulated in injected lipid nanoparticles) directed against transthyretin, led to a 56% to 67% reduction in transthyretin levels by the 28th day of study, with no significant toxicity. This trial established proof of concept for RNAi treatment of an inherited disease, an approach that will undoubtedly be applied to other diseases where elimination of the mutant gene product is the goal.

Induction of Exon Skipping

Exon skipping refers to the use of molecular interventions to exclude an exon from a pre-mRNA that encodes a reading frame–disrupting mutation, thereby rescuing expression of the mutant gene. If the number of nucleotides in the excluded exon is a multiple of three, no frame shift will occur and, if the resulting polypeptide with the deleted amino acids retains sufficient function, a therapeutic benefit will result. The most widely studied method of inducing exon skipping is through the use of antisense oligonucleotides (ASOs), which are synthetic 15- to 35-nucleotide single-stranded molecules that can hybridize to specific corresponding sequences in a pre-mRNA (see Fig. 13-7). The clearest example of the potential of this strategy is provided by Duchenne muscular dystrophy (DMD) (see Chapter 12(Case 14).

The goal of exon skipping in DMD is to convert a DMD mutation into an in-frame counterpart that generates a functional dystrophin, just as the deletions that allow the production of a partially functioning dystrophin are associated with the milder phenotype of Becker muscular dystrophy (see Fig. 12-18). The distribution of DMD mutations is nonrandomly distributed in the gene (see Chapter 12), and thus, remarkably, the skipping of just exon 51 alone would restore the dystrophin reading frame of an estimated 13% of all DMD patients (Fig. 13-13). This exon has therefore been the major focus of exon-skipping drug development. Several clinical trials have established that ASOs that cause skipping of exon 51 can produce significant increases in the number of dystrophin-positive muscle fibers of DMD patients. Moreover, one trial demonstrated stabilization of patient walking ability, but the treatment group was small and must be studied in a larger number of subjects. Irrespective of the specific challenges posed by DMD, it will be surprising if exon-skipping strategies do not ultimately play a significant role in the therapy of some inherited disorders.

image

FIGURE 13-13 Schematic representation of exon skipping. In a patient with Duchenne muscular dystrophy (DMD) who has a deletion of exon 50, an out-of-frame transcript is generated in which exon 49 is spliced to exon 51 (A). As a result, a stop codon is generated in exon 51, which prematurely aborts dystrophin synthesis. The sequence-specific binding of the exon-internal antisense oligonucleotide PRO051 interferes with the correct inclusion of exon 51 during splicing, so that the exon is actually skipped (B). This restores the open reading frame of the transcript and allows the synthesis of a dystrophin similar to that in patients with Becker muscular dystrophy (BMD). mRNA, Messenger RNA. SeeSources & Acknowledgments.

Gene Editing

Over the last decade, molecular biologists have developed methods to introduce site-specific genomic sequence changes into the DNA of intact organisms, including primates. The correction of a mutant gene sequence in its natural DNA context, in a sufficient number of target cells, would be an ideal treatment. This new technology, termed genome editing, uses engineered endonucleases containing a DNA-binding domain that will recognize a specific sequence in the genome, such as the sequence in which a missense mutation is embedded. Subsequently, a nuclease domain creates a double-stranded break, and cellular mechanisms for homology-directed repair (HDR) then repair the break (see Chapter 4), introducing the wild-type nucleotide to replace the mutant one. The template for the HDR must be based on a matching homologous wild-type DNA template that is introduced into the target cells before editing. The most widely used editing approach at present is the clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated (Cas) 9 system, commonly referred to as CRISPR/Cas9.

In humans, genome editing offers possibilities for the correction of genetic defects in their natural genomic landscape, without the risks associated with the semirandom vector integration of some viral vectors used in gene therapy (see later section). The first clinical use of this technology was a Phase I (safety) clinical trial reported in 2014. This study took advantage of the knowledge that a naturally occurring deletion in CCR5, the gene that encodes the cell membrane coreceptor for human immunodeficiency virus (HIV), renders homozygous carriers resistant to HIV infection but does not impair CD4 T-cell function (see Chapter 9). When CD4 T cells taken from HIV-infected patients were treated with an adenoviral vector expressing a nuclease designed to generate a null allele of the CCR5 gene, and then reinfused into the patient, the CCR5 gene was “knocked out” in 11% to 28% of the CD4 T cells in these patients; the modified cells had a half-life of almost 1 year, and HIV RNA became undetectable in one of four patients who could be evaluated. This study demonstrates the great clinical potential of gene editing.

A major concern whose real dimensions are presently unknown is that the endonucleases can have off-target effects, which could cause mutations elsewhere in the genome. Nevertheless, considerable optimism is justified in thinking that this technology can be extended to the correction of mutations in the cells of individuals with genetic diseases in the future, including, for example, bone marrow stem cells for the treatment of inherited blood and immune system disorders (see later discussion).

Modification of the Somatic Genome by Transplantation

Transplanted cells retain the genotype of the donor, and consequently transplantation can be regarded as a form of gene transfer therapy because it leads to a modification of the somatic genome. There are two general indications for the use of transplantation in the treatment of genetic disease. First, cells or organs may be transplanted to introduce wild-type copies of a gene into a patient with mutations in that gene. This is the case, for example, in homozygous familial hypercholesterolemia (see Chapter 12), for which liver transplantation is an effective but high-risk procedure. The second and more common indication is for cell replacement, to compensate for an organ damaged by genetic disease (for example, a liver that has become cirrhotic in α1-antitrypsin (deficiency). Some examples of the uses of transplantation in genetic disease are provided in Table 13-4.

Stem Cell Transplantation

Stem cells are defined by two properties: (1) their ability to proliferate to form the differentiated cell types of a tissue in vivo; and (2) their ability to self-renew—that is, to form another stem cell. Embryonic stem cells, which can give rise to the whole organism, are discussed in Chapter 14.

Only three types of stem cells are in clinical use at present: hematopoietic stem cells (HSCs), which can reconstitute the blood system after bone marrow transplantation; corneal stem cells, which are used to regenerate the corneal epithelium, and skin stem cells. These cells are derived from immunologically compatible donors. The possibility that other types of stem cells will be used clinically in the future is enormous because stem cell research is one of the most active and promising areas of biomedical investigation. Although it is easy to overstate the potential of such treatment, optimism about the long-term future of stem cell therapy is justified.

Hematopoietic Stem Cell Transplantation in Nonstorage Diseases.

In addition to its extensive application in the management of cancer, HSC transplantation using bone marrow stem cells is the treatment of choice for a selected group of monogenic immune deficiency disorders, including SCID of any type. Its role in the management of genetic disease in general, however, is less certain and under careful evaluation. For example, excellent outcomes have been obtained with allogenic HSC transplantation in the treatment of children with β-thalassemia and sickle cell disease. Nevertheless, for each disease that bone marrow transplantation might benefit, its outcomes must be evaluated for many years and weighed against the results obtained with other therapies.

Hematopoietic Stem Cell Transplantation for Lysosomal Storage Diseases

Transplantation of Hematopoietic Stem Cells from Bone Marrow.

Bone marrow stem cell transplants are effective in correcting lysosomal storage in many tissues including, in some diseases, the brain, through the two mechanisms depicted in Figure 13-14. First, the transplanted cells are a source of lysosomal enzymes that can be transferred to other cells through the extracellular fluid, as discussed in Chapter 12 for I-cell disease. Because bone marrow–derived cells constitute approximately 10% of the total cell mass of the body, the quantitative impact of enzymes transferred from them may be significant. Second, the mononuclear phagocyte system in tissues is derived from bone marrow stem cells so that, after bone marrow transplantation, this system is of donor origin throughout the body. Of special note are the brain perivascular microglial cells, whose bone marrow origin may partially account for the correction of nervous system abnormalities by bone marrow transplantation in some storage disorders, as we will see next in the case of Hurler syndrome, a lysosomal storage disease due to α-l-Iduronidase deficiency.

image

FIGURE 13-14 The two major mechanisms by which bone marrow transplantation or gene transfer into bone marrow may reduce the substrate accumulation in lysosomal storage diseases. In the case of either treatment, bone marrow transplantation from an allogeneic donor (A) or genetic correction of the patient's own bone marrow stem cells by gene transfer (B), the bone marrow stem cell progeny, now expressing the relevant lysosomal enzyme, expand to repopulate the monocyte-macrophage system of the patient (mechanism 1). In addition, lysosomal enzymes are released from the bone marrow cells derived from the donor or from the genetically modified marrow cells of the patient and taken up by enzyme-deficient cells from the extracellular fluid (mechanism 2).

Bone marrow transplantation corrects or reduces the visceral abnormalities of many storage diseases. For example, a normalization or reduction in the size of the enlarged liver, spleen, and heart seen in Hurler syndrome can be achieved, and improvements in upper airway obstruction, joint mobility, and corneal clouding are also obtained. Most rewarding, however, has been the impact of transplantation on the neurological component of this disease. Patients who have good developmental indices before transplantation, and who receive transplants before 24 months of age, continue to develop cognitively after transplantation, in contrast to the inexorable loss of intellectual function that otherwise occurs. Interestingly, a gene dosage effect is manifested in the donor marrow; children who receive cells from homozygousnormal donors appear to be more likely to retain fully normal intelligence than do the recipients of heterozygous donor cells.

Transplantation of Hematopoietic Stem Cells from Placental Cord Blood.

The discovery that placental cord blood is a rich source of HSCs is beginning to make a substantial impact on the treatment of genetic disease. The use of placental cord blood has three great advantages over bone marrow as a source of transplantable HSCs. First, recipients are more tolerant of histoincompatible placental blood than of other allogeneic donor cells. Thus engraftment occurs even if as many as three HLA antigens, cell surface markers encoded by the major histocompatibility complex (see Chapter 8), are mismatched between the donor and the recipient. Second, the wide availability of placental cord blood, together with the increased tolerance of histoincompatible donor cells, greatly expands the number of potential donors for any recipient. This feature is of particular significance to patients from minority ethnic groups, for whom the pool of potential donors is relatively small. Third, the risk for graft-versus-host disease is substantially reduced with use of placental cord blood cells. Cord blood transplantation from unrelated donors appears to be as effective as bone marrow transplantation from a matched donor for the treatment of Hurler syndrome (Fig. 13-15).

image

FIGURE 13-15 Preservation of neurocognitive development in children with Hurler syndrome treated by cord blood transplantation. The figure displays the mean cognitive growth curve for transplanted patients compared with unaffected children. The thin black lines represent the 95% confidence interval for transplanted patients. SeeSources & Acknowledgments.

Liver Transplantation

For some metabolic liver diseases, liver transplantation is the only treatment of known benefit. For example, the chronic liver disease associated with CF or α1AT deficiency can be treated only by liver transplantation, and together these two disorders account for a large fraction of all the liver transplants performed in the pediatric population. Liver transplantation has now been undertaken for more than two dozen genetic diseases. At present, the 5-year survival rate of all children who receive liver transplants is in the range of 70% to 85%. For almost all of these patients, the quality of life is generally much improved, the specific metabolic abnormality necessitating the transplant is corrected, and in those conditions in which hepatic damage has occurred (such as α1AT deficiency), the provision of healthy hepatic tissue restores growth and normal pubertal development.

The Problems and the Future of Transplantation

Two major problems limit the wider use of transplantation for the treatment of genetic disease. First, the mortality after transplantation is still significant, and the morbidity from superimposed infection due to the requirement for immunosuppression and graft-versus-host disease is substantial. Nevertheless, the ultimate goal of transplantation research—transplantation without immunosuppression—comes incrementally closer. The increased tolerance of the recipient to cord blood transplants, compared with bone marrow–derived donor cells, exemplifies the advances in this area.

The second problem with transplantation is the finite supply of organs, cord blood being a singular exception. For example, for all indications, including genetic disease, more than 6000 liver transplants are performed annually in the United States alone, but more than double that number are added to the waiting list each year. In addition, it remains to be demonstrated that transplanted organs are generally capable of functioning normally for a lifetime.

One solution to these difficulties involves the combination of stem cell and either genome editing or gene therapy. Here, a patient's own stem cells would be cultured in vitro and either transfected by gene therapy with the gene of interest or corrected by CRISPR/Cas9 editing and returned to the patient to repopulate the affected tissue with genetically restored cells. The identification of stem cells in a variety of adult human tissues and recent advances in gene transfer therapy offer great hope for this strategy.

Induced Pluripotent Stem Cells.

The recently developed ability to induce the formation of pluripotent stem cells (iPSCs) from somatic cells has the potential to provide the optimal solution to both of the challenges of transplantation posed earlier. In this approach somatic cells, such as skin fibroblasts, would be taken from a patient in need of a transplant, and induced to form differentiated cells of the organ of interest. For example, the loss-of-function mutation in the α1-antitrypsin gene in the fibroblasts cultured from a patient with α1AT deficiency (see Chapter 12) could be corrected, either by gene editing (see earlier section) or gene therapy (see later section); the corrected cells could then be induced to form liver-specific iPSCs, which could then be transplanted into the liver of the patient to differentiate into hepatocytes. Alternatively, mature hepatocytes derived in vitro from the genetically corrected iPSCs could be transplanted. The great merit of this approach is that the genetically corrected liver cells are derived from the patient's own genome, thus evading immunological rejection of the transplanted cells as well as graft-versus-host disease. Experimental work in animal models has established that this strategy is capable of correcting inherited disorders. Substantial hurdles with iPSCs must first be overcome, however, including establishing the safety of transplanting cells derived by iPSC methodology and preventing epigenetic modifications in the derived cell type that are not characteristic of wild-type cells of the tissue of interest.

Gene Therapy

Gene therapy is the introduction of a biologically active gene into a cell to achieve a therapeutic benefit. In 2012, the first gene therapy product was licensed in the United States and Europe for the treatment of lipoprotein lipase deficiency, and gene therapy has now been shown to be effective or extremely promising in clinical trials for almost a dozen inherited diseases, some of which are outlined in Table 13-5. These recent successes firmly establish that the treatment of genetic disease at its most fundamental level—the gene—will be increasingly feasible. The goal of gene therapy is to transfer the therapeutic gene early enough in the life of the patient to prevent the pathogenetic events that damage cells. Moreover, correction of the reversible features of genetic diseases should also be possible for many conditions.

TABLE 13-5

Examples of Inherited Diseases Treated by Gene Therapy of Somatic Tissues

image

ADA, Adenosine deaminase; Hb, hemoglobin; IV, intravenous; PEG, polyethylene glycol; SCID, severe combined immunodeficiency; WAS, Wiskott-Aldrich syndrome.

In this section, we outline the potential, methods, and probable limitations of gene transfer for the treatment of human genetic disease. The minimal requirements that must be met before the use of gene transfer can be considered for the treatment of a genetic disorder are presented in the Box.

Essential Requirements of Gene Therapy for an Inherited Disorder

• Identity of the molecular defect

The identity of the affected gene must be known.

• A functional copy of the gene

A complementary DNA (cDNA) clone of the gene or the gene itself must be available. If the gene or cDNA is too large for the current generation of vectors, a functional version of the gene from which nonessential components have been removed to reduce its size may suffice.

• An appropriate vector

The most commonly used vectors at present are derived from the adeno-associated viruses (AAVs) or retroviruses, including lentivirus.

• Knowledge of the pathophysiological mechanism

Knowledge of the pathophysiological mechanism of the disease must be sufficient to suggest that the gene transfer will ameliorate or correct the pathological process and prevent, slow, or reverse critical phenotypic abnormalities. Loss-of-function mutations require replacement with a functional gene; for diseases due to dominant negative alleles, inactivation of the mutant gene or its products will be necessary.

• Favorable risk-to-benefit ratio

A substantial disease burden and a favorable risk-to-benefit ratio, in comparison with alternative therapies, must be present.

• Appropriate regulatory components for the transferred gene

Tight regulation of the level of gene expression is relatively unimportant in some diseases and critical in others. In thalassemia, for example, overexpression of the transferred gene would cause a new imbalance of globin chains in red blood cells, whereas low levels of expression would be ineffective. In some enzymopathies, a few percent of normal expression may be therapeutic, and abnormally high levels of expression may have no adverse effect.

• An appropriate target cell

Ideally, the target cell must have a long half-life or good replicative potential in vivo. It must also be accessible for direct introduction of the gene or, alternatively, it must be possible to deliver sufficient copies of the gene to it (e.g., through the bloodstream) to attain a therapeutic benefit. The feasibility of gene therapy is often enhanced if the target cell can be cultured in vitro to facilitate gene transfer into it; in this case, it must be possible to introduce a sufficient number of the recipient cells into the patient and have them functionally integrate into the relevant organ.

• Strong evidence of efficacy and safety

Cultured cell and animal studies must indicate that the vector and gene construct are both effective and safe. The ideal precedent is to show that the gene therapy is effective, benign, and enduring in a large animal genetic model of the disease in question. At present, however, large animal models exist for only a few monogenic diseases. Genetically engineered or spontaneous mutant mouse models are much more widely available.

• Regulatory approval

Protocol review and approval by an institutional review board are essential. In most countries, human gene therapy trials are also subject to oversight by a governmental agency.

General Considerations for Gene Therapy

In the treatment of inherited disease, the most common use of gene therapy will be the introduction of functional copies of the relevant gene into the appropriate target cells of a patient with a loss-of-function mutation (because most genetic diseases result from such mutations).

In these instances, precisely where the transferred gene inserts into the genome of a cell would, in principle, generally not be important (see later discussion). If gene editing (see earlier discussion and Table 13-4) to treat inherited disease becomes possible, then correction of the defect in the mutant gene in its normal genomic context would be ideal and would alleviate concerns such as the activation of a nearby oncogene by the regulatory activity of a viral vector, or the inactivation of a tumor suppressor due to insertional mutagenesis by the vector. In some long-lived types of cells, stable, long-term expression may not require integration of the introduced gene into the host genome. For example, if the transferred gene is stabilized in the form of an episome (a stable nuclear but nonchromosomal DNA molecule, such as that formed by an adeno-associated viral vector, discussed later), and if the target cell is long-lived (e.g., T cells, neurons, myocytes, hepatocytes), then long-term expression can occur without integration.

Gene therapy may also be undertaken to inactivate the product of a dominant mutant allele whose abnormal product causes the disease. For example, vectors carrying siRNAs (see earlier section) could, in principle, be used to mediate the selective degradation of a mutant mRNA encoding a dominant negative proα1(I) collagen that causes osteogenesis imperfecta (see Chapter 12).

Gene Transfer Strategies

An appropriately engineered gene may be transferred into target cells by one of two general strategies (Fig. 13-16). The first involves introduction of the gene into cells that have been cultured from the patient ex vivo (that is, outside the body) and then reintroduction of the cells to the patient after the gene transfer. In the second approach, the gene is injected directly in vivo into the tissue or extracellular fluid of interest (from which it is taken up by the target cells). In some cases, it may be desirable to target the vector to a specific cell type; this is usually achieved by modifying the coat of a viral vector so that only the designated cells bind the viral particles.

image

FIGURE 13-16 The two major strategies used to transfer a gene to a patient. For patients with a genetic disease, the most common approach is to construct a viral vector containing the human complementary DNA (cDNA) of interest and to introduce it directly into the patient or into cells cultured from the patient that are then returned to the patient. The viral components at the ends of the molecule are required for the integration of the vector into the host genome. In some instances, the gene of interest is placed in a plasmid, which is then used for the gene transfer.

The Target Cell

The ideal target cells are stem cells (which are self-replicating) or progenitor cells taken from the patient (thereby eliminating the risk for graft-versus-host disease); both cell types have substantial replication potential. Introduction of the gene into stem cells can result in the expression of the transferred gene in a large population of daughter cells. At present, bone marrow is the only tissue whose stem cells have been successfully targeted as recipients of transferred genes. Genetically modified bone marrow stem cells have been used to cure two forms of SCID, as discussed later. Gene transfer therapy into blood stem cells is also likely to be effective for the treatment of hemoglobinopathies and storage diseases for which bone marrow transplantation has been effective, as discussed earlier.

An important logistical consideration is the number of cells into which the gene must be introduced in order to have a significant therapeutic effect. To treat PKU, for example, the approximate number of liver cells into which the phenylalanine hydroxylase gene would have to be transferred is approximately 5% of the hepatocyte mass, or approximately 1010 cells, although this number could be much less if the level of expression of the transferred gene is higher than wild type. A much greater challenge is gene therapy for muscular dystrophies, for which the gene must be inserted into a significant fraction of the huge number of myocytes in the body in order to have therapeutic efficacy.

DNA Transfer into Cells: Viral Vectors

The ideal vector for gene therapy would be safe, readily made, and easily introduced into the appropriate target tissue, and it would express the gene of interest for life. Indeed, no single vector is likely to be satisfactory in all respects for all types of gene therapy, and a repertoire of vectors will probably be required. Here, we briefly review three of the most widely used classes of viral vectors, those derived from retroviruses, adeno-associated viruses (AAVs), and adenoviruses.

One of the most widely used classes of vectors is derived from retroviruses, simple RNA viruses that can integrate into the host genome. They contain only three structural genes, which can be removed and replaced with the gene to be transferred (see Fig. 13-16). The current generation of retroviral vectors has been engineered to render them incapable of replication. In addition, they are nontoxic to the cell, and only a low number of copies of the viral DNA (with the transferred gene) integrate into the host genome. Moreover, the integrated DNA is stable and can accommodate up to 8 kb of added DNA, commodious enough for many genes that might be transferred. A major limitation of many retroviral vectors, however, is that the target cell must undergo division for integration of the virus into the host DNA, limiting the use of such vectors in nondividing cells such as neurons. In contrast, lentiviruses, the class of retroviruses that includes HIV, are capable of DNA integration in nondividing cells, including neurons. Lentiviruses have the additional advantage of not showing preferential integration into any specific gene locus, thus reducing the chances of activating an oncogene in a large number of cells.

AAVs do not elicit strong immunological responses, a great advantage that enhances the longevity of their expression. Moreover, they infect dividing or nondividing cells to remain in a predominantly episomal form that is stable and confers long-term expression of the transduced gene. A disadvantage is that the current AAV vectors can accommodate inserts of up to only 5 kb, which is smaller than many genes in their natural context.

The third group of viral vectors, adenovirus-derived vectors, can be obtained at high titer, will infect a wide variety of dividing or nondividing cell types, and can accommodate inserts of 30 to 35 kb. However, in addition to other limitations, they have been associated with at least one death in a gene therapy trial through the elicitation of a strong immune response. At present their use is restricted to gene therapy for cancer.

Risks of Gene Therapy

Gene therapy for the treatment of human disease has risks of three general types:

• Adverse response to the vector or vector-disease combination. Principal among the concerns is that the patient will have an adverse reaction to the vector or the transferred gene. Such problems should be largely anticipated with appropriate animal and preliminary human studies.

• Insertional mutagenesis causing malignancy. The second concern is insertional mutagenesis, that is, that the transferred gene will integrate into the patient's DNA and activate a proto-oncogene or disrupt a tumor suppressor gene, leading possibly to cancer (see Chapter 15). The illicit expression of an oncogene is less likely to occur with the current generation of viral vectors, which have been altered to minimize the ability of their promoters to activate the expression of adjacent host genes. Insertional inactivation of a tumor suppressor gene is likely to be infrequent and, as such, is an acceptable risk in diseases for which there is no therapeutic alternative.

• Insertional inactivation of an essential gene. A third risk—that insertional inactivation could disrupt a gene essential for viability—will, in general, be without significant effect because such lethal mutations are expected to be rare and will kill only single cells. Although vectors appear to somewhat favor insertion into transcribed genes, the chance that the same gene will be disrupted in more than a few cells is extremely low. The one exception to this statement applies to the germline; an insertion into a gene in the germline could create a dominant disease-causing mutation that might manifest in the treated patient's offspring. Such events, however, are likely to be rare and the risk acceptable because it would be difficult to justify withholding, on this basis, carefully planned and reviewed trials of gene therapy from patients who have no other recourse. Moreover, the problem of germline modification by disease treatment is not confined to gene therapy. For example, most chemotherapy used in the treatment of malignant disease is mutagenic, but this risk is accepted because of the therapeutic benefits.

Diseases That Have Been Amenable to Gene Therapy

Although nearly a dozen single-gene diseases have been shown to improve with gene therapy, a large number of other monogenic disorders are potential candidates for this strategy, including retinal degenerations; hematopoietic conditions, such as sickle cell anemia and thalassemia; and disorders affecting liver proteins, such as PKU, urea cycle disorders, familial hypercholesterolemia, and α1AT deficiency. Here we discuss several disorders in which gene therapy has been clearly effective, but which also highlight some of the challenges associated with this therapeutic approach.

Severe X-Linked Combined Immunodeficiency

The SCIDs are due to mutations in genes required for lymphocyte maturation. Affected individuals fail to thrive and die early in life of infection because they lack functional B and T lymphocytes. The most common form of the disease, X-linked SCID, results from mutations in the X-linked gene (IL2RG) encoding the γc-cytokine receptor subunit of several interleukin receptors. The receptor deficiency causes an early block in T- and natural killer–lymphocyte growth, survival, and differentiation and is associated with severe infections, failure to thrive, and death in infancy or early childhood if left untreated. This condition was chosen for a gene therapy trial for two principal reasons. First, bone marrow transplantation cures the disease, indicating that the restoration of lymphocyte expression of IL2RG can reverse the pathophysiological changes. Second, it was believed that so-called transduced cells carrying the transferred gene would have a selective survival advantage over untransduced cells.

The outcome of trials of X-linked SCID has been dramatic and resulted, in 2000, in the first gene therapy cure of a patient with a genetic disease. Subsequent confirmation has been obtained in most patients in subsequent clinical trials (see Table 13-5). Bone marrow stem cells from the patients were infected in culture (ex vivo) with a retroviral vector that expressed the γc cytokine subunit cDNA. A selective advantage was conferred on the transduced cells by the gene transfer. Transduced T cells and natural killer cells populated the blood of treated patients, and the T cells appeared to behave normally. Although the frequency of transduced B cells was low, adequate levels of serum immunoglobulin and antibody levels were obtained. Dramatic clinical improvement occurred, with resolution of protracted diarrhea and skin lesions and restoration of normal growth and development. These initial trials demonstrated the great potential of gene therapy for the correction of inherited disease.

This highly promising outcome, however, came at the cost of induction of a leukemia-like disorder in 5 of the 20 treated patients, who developed an extreme lymphocytosis resembling T-cell acute lymphocytic leukemia; 4 of them are now well after treatment of the leukemia. The malignancy was due to insertional mutagenesis: the retroviral vector inserted into the LMO2 locus, causing aberrant expression of the LMO2mRNA, which encodes a component of a transcription factor complex that mediates hematopoietic development. Consequently, trials using integrating vectors in hematopoietic cells must now monitor insertion sites and survey for clonal proliferation. Current-generation vectors are designed to avoid this mutagenic effect by using strategies such as including a self-inactivating or “suicide” gene cassette in the vector to eliminate clones of malignant cells. At this point, bone marrow stem cell transplantation remains the treatment of choice for those children with SCID fortunate enough to have a donor with an HLA-identical match. For patients without such a match, autologous transplantation of hematopoietic stem and progenitor cells, in which the genetic defect has been corrected by gene therapy, offers a lifesaving alternative, but one that may not be without risk.

Metachromatic Leukodystrophy

Metachromatic leukodystrophy (MLD) is an autosomal recessive neurodegenerative disorder that, in the late infantile form, is generally fatal by 5 years of age. It results from mutations in the gene, ARSA, that encodes arylsulfatase A, a lysosomal enzyme that degrades sulfatides that are neurotoxic, leading to demyelination in the central and peripheral nervous system. As described earlier, HSC transplantation is an effective treatment of some lysosomal storage diseases because some of the donor-derived macrophages and microglia can enter the central nervous system, scavenge the stored material (such as sulfatide in MLD), and release lysosomal enzymes that are taken up by the mutant cells of the patient. HSC transplants have not been successful for MLD, however, a failure thought to be due to a level of ARSA expression from the transplanted cells that is too low to have a therapeutic effect.

In an apparently successful treatment, the autologous HSCs of three patients with MLD were transduced with a lentiviral vector that was engineered to produce above-normal levels of arylsulfatase A from a functional ARSA gene, and the genetically corrected HSCs were then engrafted (Fig. 13-17). Although more than 36,000 different lentiviral integration sites were examined, no evidence of genotoxicity was observed, suggesting that lentiviral vectors can be effective in the gene therapy of HSCs. Dramatically, disease progression was arrested, at least up to 24 months after treatment, but long-term follow-up will be required to establish that the effect of the gene therapy is benign and enduring.

image

FIGURE 13-17 Clinical follow-up of a metachromatic leukodystrophy (MLD) patient after hematopoietic stem cell gene therapy (GT) with the arylsulfatase A gene. Magnetic resonance images from patient MLD01 before gene therapy and 2 years after treatment. The brain of this patient appeared largely normal 2 years after treatment. In contrast, the brain of an untreated, age matched late infantile MLD patient (UT LI MLD) showed severe demyelination associated with diffuse atrophy. In MLD01 images, a small area of hyperintensity is present within the splenium of the corpus callosum (white arrow). This area appeared at the 12-month follow-up and remained stable thereafter. In UT LI MLD images, extensive, diffuse symmetrical hyperintensities with typical striped “tigroid pattern” (white arrows) are seen within periventricular white matter, corpus callosum, external and internal capsules, and cerebellar deep white matter. Severe diffuse brain atrophy involving basal ganglia and thalamus, which show a T2 hypointense signal, is also present. SeeSources & Acknowledgments.

Hemophilia B

Hemophilia B is an X-linked disorder of coagulation caused by mutations in the F9 gene, leading to a deficiency or dysfunction of clotting factor IX (Case 21). The disease is characterized by bleeding into soft tissues, muscles, and weight-bearing joints, and occurs within hours to days after trauma. Severely affected subjects, with less than 1% of normal levels of factor IX, have frequent bleeding that causes crippling joint disease and early death. Prophylactic—but not curative—treatment with intravenous factor IX concentrate several times a week is expensive and leads to the generation of inhibitory antibodies.

In 2011, the first successful gene therapy treatment of hemophilia B was reported in six patients using an AAV8 vector that is tropic for hepatocytes, where factor IX is normally produced. After a single infusion of the AAV8-F9 vector, four patients were able to discontinue prophylactic factor IX infusions, whereas the other two tolerated longer intervals between infusions. The two patients who received the highest dose of the vector had transient asymptomatic increases in liver enzyme levels—which resolved with steroid treatment—indicating that immune-related side effects must remain a concern in future studies. Unfortunately, the AAV vectors cannot accommodate the gene for factor VIII, so that other vectors will have to be developed for hemophilia A patients. Apart from this limitation of cargo size, however, AAV-mediated gene therapy targeted to hepatocytes may be applicable to any genetic disease in which production of the protein in the liver is the desired goal.

β-Thalassemia

The hemoglobinopathies are the most common genetic defects in the world (see Chapter 11), but at present they are incurable except by HSC transplantation from a matched donor. Consequently, the development of effective, safe, and affordable gene therapy for these disorders, the most common being sickle cell disease and the α- and β-thalassemias, would be a medical triumph.

In 2010, the first successful gene therapy trial for a hemoglobinopathy was reported, in a single patient with β-thalassemia who was transfusion-dependent, with hemoglobin levels of only 4 to 6 g/dL. This individual was a genetic compound of βE and β0 alleles, the βE allele generating a mutant β-globin of decreased abundance, with the β0 allele being a null. The patient's HSCs were transduced with a lentiviral vector containing a β-globin gene. The patient became transfusion-independent, with hemoglobin levels ranging from 9 to 10 g/dL, although the vector-encoded hemoglobin accounted for only approximately one third of the total, the remainder being the mutant Hb E and Hb F. Unexpectedly, the increase in normal β-globin expression was largely attributable to one bone marrow cell clone, in which the lentiviral vector integrated into a gene encoding a transcriptional regulator called HMGA2. This integration activated expression in erythroid cells of a truncated form of HMGA2, an event that confounded the interpretation of the result, because the extent to which the clonal dominance of cells expressing the truncated HMGA2 accounted for the therapeutic benefits of the gene therapy is unclear.

This study offers great promise but highlights the potential risks associated with the random insertion of viral vectors in the genome. Much current research is therefore devoted to the development of safer gene delivery vectors, including modified lentiviral vectors.

The Prospects for Gene Therapy

To date, almost 2000 clinical gene therapy trials (approximately two thirds of which are for cancer) have been undertaken worldwide to evaluate both the safety and efficacy of this long-promised and conceptually promising technology. Approximately 180 of these trials were for the treatment of monogenic diseases. The exciting results obtained with gene therapy to date, albeit with small numbers of patients and only a few diseases, validates the optimism behind this immense effort. Although the breadth of applications remains uncertain, it is to be hoped that over the next few decades, gene therapy for both monogenic and genetically complex diseases will contribute to the management of many disorders, both common and rare.

Precision Medicine: the Present and Future of the Treatment of Mendelian Disease

The treatment of single-gene diseases embodies the concept of precision medicine tailored to the individual patient as deeply as any other area of medical treatment. Knowledge of the specific mutant sequence in an individual is central to many of the targeted therapies described in this chapter. The promise of gene therapy for an individual with a mendelian disorder must be based on the identification of the mutant gene in each affected individual and on the design of a vector that will deliver the therapeutic gene to the targeted tissue. Similarly, approaches based on gene editing require knowledge of the specific mutation to be corrected.

Beyond this, however, precision medicine will frequently require knowledge of the precise mutant allele and of its specific effect on the mRNA and protein. In many cases, the exact nature of the mutation will define the drug that will bind to a specific regulatory sequence to enhance or reduce the expression of a gene. In other cases, the mutation will dictate the sequence of an allele-specific oligonucleotide to mediate the skipping of an exon with a premature termination codon, or of an siRNA to suppress a dominant negative allele. A compendium of small molecules will gradually become available to suppress particular stop codons, to act as chaperones that will rescue mutant proteins from misfolding and proteosomal degradation, or to potentiate the activity of mutant proteins.

Genetic treatment is not only becoming more and more creative, it is becoming more and more precise. The future promises not only a longer life for many patients, but a life of vastly better quality.

General References

Campeau PM, Scriver CR, Mitchell JJ. A 25-year longitudinal analysis of treatment efficacy in inborn errors of metabolism. Mol Genet Metab. 2008;95:11–16.

Dietz HC. New therapeutic approaches to mendelian disorders. N Engl J Med. 2010;363:852–863.

Valle D, Beaudet AL, Vogelstein B, et al. The online metabolic and molecular bases of inherited disease. 2014 [Available at] http://ommbid.mhmedical.com/book.aspx?bookID=474.

References for Specific Topics

Arora N, Daley GQ. Pluripotent stem cells in research and treatment of hemoglobinopathies. Cold Spring Harb Perspect Med. 2012;2:a011841.

Bélanger-Quintana A, Burlina A, Harding CO, et al. Up to date knowledge on different treatment strategies for phenylketonuria. Mol Genet Metabolism. 2011;104:S19–S25.

Biffi A, Montini E, Lorioli L, et al. Lentiviral hematopoietic stem cell gene therapy benefits metachromatic leukodystrophy. Science. 2013;341:1233158; 10.1126/science.1233158.

Cathomen T, Ehl S. Translating the genomic revolution—targeted genome editing in primates. N Engl J Med. 2014;370:2342–2345.

Coelho T, Adams D, Silva A, et al. Safety and efficacy of RNAi therapy for transthyretin amyloidosis. N Engl J Med. 2013;369(9):818–829.

Daley GQ. The promise and perils of stem cell therapeutics. Cell Stem Cell. 2012;10:740–749.

Desnick RJ, Schuchman EH. Enzyme replacement therapy for lysosomal diseases: lessons from 20 years of experience and remaining challenges. Annu Rev Genomics Hum Genet. 2012;13:307–335.

de Souza N. Primer: genome editing with engineered nucleases. Nat Methods. 2012;9:27.

Dong A, Rivella S, Breda L. Gene therapy for hemoglobinopathies: progress and challenges. Trans Res. 2013;161:293–306.

Gaspar HB, Qasim W, Davies EG, et al. How I treat severe combined immunodeficiency. Blood. 2013;122:3749–3758.

Gaziev J, Lucarelli G. Hematopoietic stem cell transplantation for thalassemia. Curr Stem Cell Res Ther. 2011;6:162–169.

Goemans NM, Tulinius M, van den Akker JT. Systemic administration of PRO051 in Duchenne's muscular dystrophy. N Engl J Med. 2011;364:1513–1522.

Groenink M, den Hartog AW, Franken R, et al. Losartan reduces aortic dilatation rate in adults with Marfan syndrome: a randomized controlled trial. Eur Heart J. 2013;34:3491–3500.

Hanna JH, Saha K, Jaenisch R. Pluripotency and cellular reprogramming: facts, hypotheses, unresolved issues. Cell. 2010;143:508–525.

Hanrahan JW, Sampson HM, Thomas DY. Novel pharmacological strategies to treat cystic fibrosis. Trends Pharmacol Sci. 2013;34:119–125.

High KA. Gene therapy in clinical medicine. Longo D, Fauci A, Kasper D, et al. Harrison's principles of internal medicine. ed 19. McGraw-Hill: New York; 2015 [in press].

Huang R, Southall N, Wang Y, et al. The NCGC Pharmaceutical Collection: A comprehensive resource of clinically approved drugs enabling repurposing and chemical genomics. Sci Transl Med. 2011;3:80ps16.

Jarmin S, Kymalainen H, Popplewell L, et al. New developments in the use of gene therapy to treat Duchenne muscular dystrophy. Expert Opin Biol Ther. 2014;14:209–230.

Johnson SM, Connelly S, Fearns C, et al. The transthyretin amyloidoses: from delineating the molecular mechanism of aggregation linked to pathology to a regulatory agency approved drug. J Mol Biol. 2012;421:185–203.

Li M, Suzuki K, Kim NY, et al. A cut above the rest: targeted genome editing technologies in human pluripotent stem cells. J Biol Chem. 2014;289:4594–4599.

Mukherjee S, Thrasher AJ. Gene therapy for primary immunodeficiency disorders: progress, pitfalls and prospects. Gene. 2013;525:174–181.

Nathwani AC, Tuddenham EGD, Rangarajan S. Adenovirus-associated virus vector–mediated gene transfer in hemophilia B. N Engl J Med. 2011;365:2357–2365.

Okam MM, Ebert BL. Novel approaches to the treatment of sickle cell disease: the potential of histone deacetylase inhibitors. Expert Rev Hematol. 2012;5:303–311.

Otsuru S, Gordon PL, Shimono K, et al. Transplanted bone marrow mononuclear cells and MSCs impart clinical benefit to children with osteogenesis imperfecta through different mechanisms. Blood. 2012;120:1933–1941.

Peltz SW, Morsy M, Welch EW, et al. Ataluren as an agent for therapeutic nonsense suppression. Annu Rev Med. 2013;64:407–425.

Perrine SP, Pace BS, Faller DV. Targeted fetal hemoglobin induction for treatment of beta hemoglobinopathies. Hematol Oncol Clin North Am. 2014;28:233–248.

Prasad VK, Kurtzberg J. Cord blood and bone marrow transplantation in inherited metabolic diseases: scientific basis, current status and future directions. Br J Haematol. 2009;148:356–372.

Ramsey BW, Davies J, McElvaney NG, et al. A CFTR potentiator in patients with cystic fibrosis and the G551D mutation. N Engl J Med. 2011;365:1663–1672.

Robinton DA, Daley GQ. The promise of induced pluripotent stem cells in research and therapy. Nature. 2012;481:295–305.

Sander JD, Joung JK. CRISPR-Cas systems for editing, regulating and targeting genomes. Nat Biotechnol. 2014;32:347–355.

Southwell AL, Skotte NH, Bennett CF, et al. Antisense oligonucleotide therapeutics for inherited neurodegenerative diseases. Trends Mol Med. 2012;18:634–643.

Tebas P, Stein D, Tang WW, et al. Gene editing of CCR5 in autologous CD4 T cells of persons infected with HIV. N Engl J Med. 2014;370:901–910.

van Ommen G-JB, Aartsma-Rus A. Advances in therapeutic RNA-targeting. Trends Mol Med. 2012;18:634–643.

Verma IM. Gene therapy that works. Science. 2013;341:853–855.

Xu J, Peng C, Sankaran VG, et al. Correction of sickle cell disease in adult mice by interference with fetal hemoglobin silencing. Science. 2011;334:993–996.

Useful Websites

Registry and results database of publicly and privately supported clinical studies of human participants conducted around the world: https://clinicaltrials.gov/.

Gene Therapy Clinical Trials Worldwide: http://www.wiley.com/legacy/wileychi/genmed/clinical/.

Problems

1. X-linked chronic granulomatous disease (CGD) is an uncommon disorder characterized by a defect in host defense that leads to severe, recurrent, and often fatal pyogenic infections beginning in early childhood. The X-linked CGD locus encodes the heavy chain of cytochrome b, a component of the oxidase that generates superoxide in phagocytes. Because interferon-γ (IFN-γ) is known to enhance the oxidase activity of normal phagocytes, IFN-γ was administered to boys with X-linked CGD to see whether their oxidase activity increased. Before treatment, the phagocytes of some less severely affected patients had small but detectable bursts of oxidase activity (unlike those of severely affected patients), suggesting that increased activity in these less severely affected subjects is the result of greater production of cytochrome b from the affected locus. In these less severe cases, IFN-γ increased the cytochrome b content, superoxide production, and killing of Staphylococcus aureus in the granulocytes. The IFN-γ effect was associated with a definite increase in the abundance of the cytochrome b chain. Presumably, the cytochrome b polypeptide of these patients is partially functional, and increased expression of the residual function improved the physiological defect. Describe the genetic differences that might account for the fact that the phagocytes of some patients with X-linked CGD respond to IFN-γ in vitro and others do not.

2. Identify some of the limitations on the types of proteins that can be considered for extracellular replacement therapy, as exemplified by polyethylene glycol–adenosine deaminase (PEG-ADA). What makes this approach inappropriate for phenylalanine hydroxylase deficiency? If Tay-Sachs disease caused only liver disease, would this strategy succeed? If not, why?

3. A 3-year-old girl, Rhonda, has familial hypercholesterolemia due to a deletion of the 5′ end of each of her low-density lipoprotein (LDL) receptor genes that removed the promoter and the first two exons. (Rhonda's parents are second cousins.) You explain to the parents that she will require plasmapheresis every 1 to 2 weeks for years. At the clinic, however, they meet another family with a 5-year-old boy with the same disease. The boy has been treated with drugs with some success. Rhonda's parents want to know why she has not been offered similar pharmacological therapy. Explain.

4. What classes of mutations are likely to be found in homocystinuric patients who are not responsive to the administration of large doses of pyridoxine (vitamin B6)? How might you explain the fact that Tom is completely responsive, whereas his first cousin Allan has only a partial reduction in plasma homocystine levels when he is given the same amount of vitamin B6?

5. You have isolated the gene for phenylalanine hydroxylase (PAH) and wish ultimately to introduce it into patients with PKU. Your approach will be to culture cells from the patient, introduce a functional version of the gene into the cells, and reintroduce the cells into the patient.

a. What DNA components do you need to make a functional PAH protein in a gene transfer experiment?

b. Which tissues would you choose in which to express the enzyme, and why? How does this choice affect your gene construct in (a)?

c. You introduce your version of the gene into fibroblasts cultured from a skin biopsy specimen from the patient. Northern (RNA) blot analysis shows that the messenger RNA (mRNA) is present in normal amounts and is the correct size. However, no PAH protein can be detected in the cells. What kinds of abnormalities in the transferred gene would explain this finding?

d. You have corrected all the problems identified in (c). On introducing the new version of the gene into the cultured cells, you now find that the PAH protein is present in great abundance, and when you harvest the cells and assay the enzyme (in the presence of all the required components), normal activity is obtained. However, when you add 3H-labeled phenylalanine to the cells in culture, no 3H-labeled tyrosine is formed (in contrast, some cultured liver cells produce a large quantity of 3H-labeled tyrosine in this situation). What are the most likely explanations for the failure to form 3H-tyrosine? How does this result affect your gene therapy approach to patients?

e. You have developed a method to introduce your functional version of the gene directly into a large proportion of the hepatocytes of patients with PAH deficiency. Unexpectedly, you find that much lower levels of PAH enzymatic activity are obtained in patients in whom significant amounts of the inactive PAH homodimer were detectable in hepatocytes before treatment than in patients who had no detectable PAH polypeptide before treatment. How can you explain this result? How might you overcome the problem?

6. Both alleles of an autosomal gene that is mutant in your patient produce a protein that is decreased in abundance but has residual function. What therapeutic strategies might you consider in such a situation?

7. A Phase III clinical trial is undertaken to evaluate the effectiveness of a small molecule drug that facilitates skipping over nonsense mutation codons. The drug had been shown in earlier trials to have a modest but significant clinical effect in patients with cystic fibrosis with at least one CFTR nonsense mutation. Two cystic fibrosis (CF) patients each have a nonsense mutation in one CFTR allele, but at different locations in the reading frame. One patient responds to the drug, whereas the other does not. Discuss how the location of the nonsense mutation in the predicted reading frame of the protein could account for this differential response.