Basic and Clinical Pharmacology, 13th Ed.

Agents Used in Cytopenias; Hematopoietic Growth Factors

James L. Zehnder, MD*


A 65-year-old woman with a long-standing history of poorly controlled type 2 diabetes mellitus presents with increasing numbness and paresthesias in her extremities, generalized weakness, a sore tongue, and gastrointestinal discomfort. Physical examination reveals a frail-looking, pale woman with diminished vibration sensation, diminished spinal reflexes, and a positive Babinski sign. Examination of her oral cavity reveals atrophic glossitis, in which the tongue appears deep red in color and abnormally smooth and shiny due to atrophy of the lingual papillae. Laboratory testing reveals a macrocytic anemia based on a hematocrit of 30% (normal for women, 37–48%), a hemoglobin concentration of 9.4 g/dL (normal for elderly women, 11.7–13.8 g/dL), an erythrocyte mean cell volume (MCV) of 123 fL (normal, 84–99 fL), an erythrocyte mean cell hemoglobin concentration (MCHC) of 34% (normal, 31–36%), and a low reticulocyte count. Further laboratory testing reveals a normal serum folate concentration and a serum vitamin B12 (cobalamin) concentration of 98 pg/mL (normal, 250–1100 pg/mL). Results of a Schilling test indicate a diagnosis of pernicious anemia. Once megaloblastic anemia was identified, why was it important to measure serum concentrations of both folic acid and cobalamin? Should this patient be treated with oral or parenteral vitamin B12?

Hematopoiesis, the production from undifferentiated stem cells of circulating erythrocytes, platelets, and leukocytes, is a remarkable process that produces over 200 billion new blood cells per day in the normal person and even greater numbers of cells in people with conditions that cause loss or destruction of blood cells. The hematopoietic machinery resides primarily in the bone marrow in adults and requires a constant supply of three essential nutrients—iron, vitamin B12, and folic acid—as well as the presence of hematopoietic growth factors, proteins that regulate the proliferation and differentiation of hematopoietic cells. Inadequate supplies of either the essential nutrients or the growth factors result in deficiency of functional blood cells. Anemia, a deficiency in oxygen-carrying erythrocytes, is the most common and several forms are easily treated. Sickle cell anemia, a condition resulting from a genetic alteration in the hemoglobin molecule, is common but is not easily treated. It is discussed in the Box: Sickle Cell Disease and Hydroxyurea. Thrombocytopenia and neutropenia are not rare, and some forms are amenable to drug therapy. In this chapter, we first consider treatment of anemia due to deficiency of iron, vitamin B12, or folic acid and then turn to the medical use of hematopoietic growth factors to combat anemia, thrombocytopenia, and neutropenia, and to support stem cell transplantation.



Basic Pharmacology

Iron deficiency is the most common cause of chronic anemia. Like other forms of chronic anemia, iron deficiency anemia leads to pallor, fatigue, dizziness, exertional dyspnea, and other generalized symptoms of tissue hypoxia. The cardiovascular adaptations to chronic anemia—tachycardia, increased cardiac output, vasodilation—can worsen the condition of patients with underlying cardiovascular disease.

Iron forms the nucleus of the iron-porphyrin heme ring, which together with globin chains forms hemoglobin. Hemoglobin reversibly binds oxygen and provides the critical mechanism for oxygen delivery from the lungs to other tissues. In the absence of adequate iron, small erythrocytes with insufficient hemoglobin are formed, giving rise to microcytic hypochromic anemia. Iron-containing heme is also an essential component of myoglobin, cytochromes, and other proteins with diverse biologic functions.


Free inorganic iron is extremely toxic, but iron is required for essential proteins such as hemoglobin; therefore, evolution has provided an elaborate system for regulating iron absorption, transport, and storage (Figure 33–1). The system uses specialized transport, storage, ferrireductase, and ferroxidase proteins whose concentrations are controlled by the body’s demand for hemoglobin synthesis and adequate iron stores (Table 33–1). A peptide called hepcidin, produced primarily by liver cells, serves as a key central regulator of the system. Nearly all of the iron used to support hematopoiesis is reclaimed from catalysis of the hemoglobin in senescent or damaged erythrocytes. Normally, only a small amount of iron is lost from the body each day, so dietary requirements are small and easily fulfilled by the iron available in a wide variety of foods. However, in special populations with either increased iron requirements (eg, growing children, pregnant women) or increased losses of iron (eg, menstruating women), iron requirements can exceed normal dietary supplies, and iron deficiency can develop.


FIGURE 33–1 Absorption, transport, and storage of iron. Intestinal epithelial cells actively absorb inorganic iron via the divalent metal transporter 1 (DMT1) and heme iron via the heme carrier protein 1 (HCP1). Iron that is absorbed or released from absorbed heme iron in the intestine (1) is actively transported into the blood by ferroportin (FP) or complexed with apoferritin (AF) and stored as ferritin (F). In the blood, iron is transported by transferrin (Tf) to erythroid precursors in the bone marrow for synthesis of hemoglobin (Hgb) (2) or to hepatocytes for storage as ferritin (3). The transferrin-iron complex binds to transferrin receptors (TfR) in erythroid precursors and hepatocytes and is internalized. After release of iron, the TfR-Tf complex is recycled to the plasma membrane and Tf is released. Macrophages that phagocytize senescent erythrocytes (RBC) reclaim the iron from the RBC hemoglobin and either export it or store it as ferritin (4). Hepatocytes use several mechanisms to take up iron and store the iron as ferritin. FO, ferroxidase. (Reproduced, with permission, from Trevor A et al: Pharmacology Examination & Board Review, 9th ed. McGraw-Hill, 2010. Copyright © The McGraw-Hill Companies, Inc.)

TABLE 33–1 Iron distribution in normal adults.1


Sickle Cell Disease and Hydroxyurea

Sickle cell disease is an important genetic cause of hemolytic anemia, a form of anemia due to increased erythrocyte destruction, instead of the reduced mature erythrocyte production seen with iron, folic acid, and vitamin B12deficiency. Patients with sickle cell disease are homozygous for the aberrant β-hemoglobin S (HbS) allele (substitution of valine for glutamic acid at amino acid 6 of β-globin) or heterozygous for HbS and a second mutated β-hemoglobin gene such as hemoglobin C (HbC) or β-thalassemia. Sickle cell disease has an increased prevalence in individuals of African descent because the heterozygous trait confers resistance to malaria.

In the majority of patients with sickle cell disease, anemia is not the major problem; the anemia is generally well compensated even though such individuals have a chronically low hematocrit (20–30%), a low serum hemoglobin level (7–10 g/dL), and an elevated reticulocyte count. Instead, the primary problem is that deoxygenated HbS chains form polymeric structures that dramatically change erythrocyte shape, reduce deformability, and elicit membrane permeability changes that further promote hemoglobin polymerization. Abnormal erythrocytes aggregate in the microvasculature—where oxygen tension is low and hemoglobin is deoxygenated—and cause veno-occlusive damage. In the musculoskeletal system, this results in characteristic, extremely severe bone and joint pain. In the cerebral vascular system, it causes ischemic stroke. Damage to the spleen increases the risk of infection, particularly by encapsulated bacteria such as Streptococcus pneumoniae. In the pulmonary system, there is an increased risk of infection and, in adults, an increase in embolism and pulmonary hypertension. Supportive treatment includes analgesics, antibiotics, pneumococcal vaccination, and blood transfusions. In addition, the cancer chemotherapeutic drug hydroxyurea (hydroxycarbamide) reduces veno-occlusive events. It is approved in the United States for treatment of adults with recurrent sickle cell crises and approved in Europe in adults and children with recurrent vaso-occlusive events. As an anticancer drug used in the treatment of chronic and acute myelogenous leukemia, hydroxyurea inhibits ribonucleotide reductase and thereby depletes deoxynucleoside triphosphate and arrests cells in the S phase of the cell cycle (see Chapter 54). In the treatment of sickle cell disease, hydroxyurea acts through poorly defined pathways to increase the production of fetal hemoglobin γ (HbF), which interferes with the polymerization of HbS. Clinical trials have shown that hydroxyurea decreases painful crises in adults and children with severe sickle cell disease. Its adverse effects include hematopoietic depression, gastrointestinal effects, and teratogenicity in pregnant women.

A. Absorption

The average American diet contains 10–15 mg of elemental iron daily. A normal individual absorbs 5–10% of this iron, or about 0.5–1 mg daily. Iron is absorbed in the duodenum and proximal jejunum, although the more distal small intestine can absorb iron if necessary. Iron absorption increases in response to low iron stores or increased iron requirements. Total iron absorption increases to 1–2 mg/d in menstruating women and may be as high as 3–4 mg/d in pregnant women.

Iron is available in a wide variety of foods but is especially abundant in meat. The iron in meat protein can be efficiently absorbed, because heme iron in meat hemoglobin and myoglobin can be absorbed intact without first having to be dissociated into elemental iron (Figure 33–1). Iron in other foods, especially vegetables and grains, is often tightly bound to organic compounds and is much less available for absorption. Nonheme iron in foods and iron in inorganic iron salts and complexes must be reduced by a ferrireductase to ferrous iron (Fe2+) before it can be absorbed by intestinal mucosal cells.

Iron crosses the luminal membrane of the intestinal mucosal cell by two mechanisms: active transport of ferrous iron by the divalent metal transporter DMT1, and absorption of iron complexed with heme (Figure 33–1). Together with iron split from absorbed heme, the newly absorbed iron can be actively transported into the blood across the basolateral membrane by a transporter known as ferroportin and oxidized to ferric iron (Fe3+) by the ferroxidase hephaestin. The liver-derived hepcidin inhibits intestinal cell iron release by binding to ferroportin and triggering its internalization and destruction. Excess iron is stored in intestinal epithelial cells as ferritin, a water-soluble complex consisting of a core of ferric hydroxide covered by a shell of a specialized storage protein called apoferritin.

B. Transport

Iron is transported in the plasma bound to transferrin, a β-globulin that can bind two molecules of ferric iron (Figure 33–1). The transferrin-iron complex enters maturing erythroid cells by a specific receptor mechanism. Transferrin receptors—integral membrane glycoproteins present in large numbers on proliferating erythroid cells—bind and internalize the transferrin-iron complex through the process of receptor-mediated endocytosis. In endosomes, the ferric iron is released, reduced to ferrous iron, and transported by DMT1 into the cytoplasm, where it is funneled into hemoglobin synthesis or stored as ferritin. The transferrin-transferrin receptor complex is recycled to the cell membrane, where the transferrin dissociates and returns to the plasma. This process provides an efficient mechanism for supplying the iron required by developing red blood cells.

Increased erythropoiesis is associated with an increase in the number of transferrin receptors on developing erythroid cells and a reduction in hepatic hepcidin release. Iron store depletion and iron deficiency anemia are associated with an increased concentration of serum transferrin.

C. Storage

In addition to the storage of iron in intestinal mucosal cells, iron is also stored, primarily as ferritin, in macrophages in the liver, spleen, and bone, and in parenchymal liver cells (Figure 33–1). The mobilization of iron from macrophages and hepatocytes is primarily controlled by hepcidin regulation of ferroportin activity. Low hepcidin concentrations result in iron release from these storage sites; high hepcidin concentrations inhibit iron release. Ferritin is detectable in serum. Since the ferritin present in serum is in equilibrium with storage ferritin in reticuloendothelial tissues, the serum ferritin level can be used to estimate total body iron stores.

D. Elimination

There is no mechanism for excretion of iron. Small amounts are lost in the feces by exfoliation of intestinal mucosal cells, and trace amounts are excreted in bile, urine, and sweat. These losses account for no more than 1 mg of iron per day. Because the body’s ability to excrete iron is so limited, regulation of iron balance must be achieved by changing intestinal absorption and storage of iron in response to the body’s needs. As noted below, impaired regulation of iron absorption leads to serious pathology.

Clinical Pharmacology

A. Indications for the Use of Iron

The only clinical indication for the use of iron preparations is the treatment or prevention of iron deficiency anemia. This manifests as a hypochromic, microcytic anemia in which the erythrocyte mean cell volume (MCV) and the mean cell hemoglobin concentration are low (Table 33–2). Iron deficiency is commonly seen in populations with increased iron requirements. These include infants, especially premature infants; children during rapid growth periods; pregnant and lactating women; and patients with chronic kidney disease who lose erythrocytes at a relatively high rate during hemodialysis and also form them at a high rate as a result of treatment with the erythrocyte growth factor erythropoietin (see below). Inadequate iron absorption can also cause iron deficiency. This is seen after gastrectomy and in patients with severe small bowel disease that results in generalized malabsorption.

TABLE 33–2 Distinguishing features of the nutritional anemias.


The most common cause of iron deficiency in adults is blood loss. Menstruating women lose about 30 mg of iron with each menstrual period; women with heavy menstrual bleeding may lose much more. Thus, many premenopausal women have low iron stores or even iron deficiency. In men and postmenopausal women, the most common site of blood loss is the gastrointestinal tract. Patients with unexplained iron deficiency anemia should be evaluated for occult gastrointestinal bleeding.

B. Treatment

Iron deficiency anemia is treated with oral or parenteral iron preparations. Oral iron corrects the anemia just as rapidly and completely as parenteral iron in most cases if iron absorption from the gastrointestinal tract is normal. An exception is the high requirement for iron of patients with advanced chronic kidney disease who are undergoing hemodialysis and treatment with erythropoietin; for these patients, parenteral iron administration is preferred.

1. Oral iron therapyA wide variety of oral iron preparations is available. Because ferrous iron is most efficiently absorbed, ferrous salts should be used. Ferrous sulfate, ferrous gluconate, and ferrous fumarate are all effective and inexpensive and are recommended for the treatment of most patients.

Different iron salts provide different amounts of elemental iron, as shown in Table 33–3. In an iron-deficient individual, about 50–100 mg of iron can be incorporated into hemoglobin daily, and about 25% of oral iron given as ferrous salt can be absorbed. Therefore, 200–400 mg of elemental iron should be given daily to correct iron deficiency most rapidly. Patients unable to tolerate such large doses of iron can be given lower daily doses of iron, which results in slower but still complete correction of iron deficiency. Treatment with oral iron should be continued for 3–6 months after correction of the cause of the iron loss. This corrects the anemia and replenishes iron stores.

TABLE 33–3 Some commonly used oral iron preparations.


Common adverse effects of oral iron therapy include nausea, epigastric discomfort, abdominal cramps, constipation, and diarrhea. These effects are usually dose-related and can often be overcome by lowering the daily dose of iron or by taking the tablets immediately after or with meals. Some patients have less severe gastrointestinal adverse effects with one iron salt than another and benefit from changing preparations. Patients taking oral iron develop black stools; this has no clinical significance in itself but may obscure the diagnosis of continued gastrointestinal blood loss.

2. Parenteral iron therapyParenteral therapy should be reserved for patients with documented iron deficiency who are unable to tolerate or absorb oral iron and for patients with extensive chronic anemia who cannot be maintained with oral iron alone. This includes patients with advanced chronic renal disease requiring hemodialysis and treatment with erythropoietin, various postgastrectomy conditions and previous small bowel resection, inflammatory bowel disease involving the proximal small bowel, and malabsorption syndromes.

The challenge with parenteral iron therapy is that parenteral administration of inorganic free ferric iron produces serious dose-dependent toxicity, which severely limits the dose that can be administered. However, when the ferric iron is formulated as a colloid containing particles with a core of iron oxyhydroxide surrounded by a core of carbohydrate, bioactive iron is released slowly from the stable colloid particles. In the United States, the three traditional forms of parenteral iron are iron dextransodium ferric gluconate complex, and iron sucrose. Two newer preparations are available (see below).

Iron dextran is a stable complex of ferric oxyhydroxide and dextran polymers containing 50 mg of elemental iron per milliliter of solution. It can be given by deep intramuscular injection or by intravenous infusion, although the intravenous route is used most commonly. Intravenous administration eliminates the local pain and tissue staining that often occur with the intramuscular route and allows delivery of the entire dose of iron necessary to correct the iron deficiency at one time. Adverse effects of intravenous iron dextran therapy include headache, light-headedness, fever, arthralgias, nausea and vomiting, back pain, flushing, urticaria, bronchospasm, and, rarely, anaphylaxis and death. Owing to the risk of a hypersensitivity reaction, a small test dose of iron dextran should always be given before full intramuscular or intravenous doses are given. Patients with a strong history of allergy and patients who have previously received parenteral iron dextran are more likely to have hypersensitivity reactions after treatment with parenteral iron dextran. The iron dextran formulations used clinically are distinguishable as high-molecular-weight and low-molecular-weight forms. In the United States, the InFeD preparation is a low-molecular-weight form while DexFerrum is a high-molecular-weight form. Clinical data—primarily from observational studies—indicate that the risk of anaphylaxis is largely associated with high-molecular-weight formulations.

Sodium ferric gluconate complex and iron-sucrose complex are alternative parenteral iron preparations. Ferric carboxymaltose is a colloidal iron preparation embedded within a carbohydrate polymer. Ferumoxytol is a superparamagnetic iron oxide nanoparticle coated with carbohydrate. The carbohydrate shell is removed in the reticuloendothelial system, allowing the iron to be stored as ferritin, or released to transferrin. Ferumoxytol may interfere with magnetic resonance imaging (MRI) studies. Thus if imaging is needed, MRI should be performed prior to ferumoxytol therapy or alternative imaging modality used if needed soon after dosing.

These agents can be given only by the intravenous route. They appear to be less likely than high-molecular-weight iron dextran to cause hypersensitivity reactions.

For patients treated chronically with parenteral iron, it is important to monitor iron storage levels to avoid the serious toxicity associated with iron overload. Unlike oral iron therapy, which is subject to the regulatory mechanism provided by the intestinal uptake system, parenteral administration—which bypasses this regulatory system—can deliver more iron than can be safely stored. Iron stores can be estimated on the basis of serum concentrations of ferritin and the transferrin saturation, which is the ratio of the total serum iron concentration to the total iron-binding capacity (TIBC).

Clinical Toxicity

A. Acute Iron Toxicity

Acute iron toxicity is seen almost exclusively in young children who accidentally ingest iron tablets. As few as 10 tablets of any of the commonly available oral iron preparations can be lethal in young children. Adult patients taking oral iron preparations should be instructed to store tablets in child-proof containers out of the reach of children. Children who are poisoned with oral iron experience necrotizing gastroenteritis with vomiting, abdominal pain, and bloody diarrhea followed by shock, lethargy, and dyspnea. Subsequently, improvement is often noted, but this may be followed by severe metabolic acidosis, coma, and death. Urgent treatment is necessary. Whole bowel irrigation(see Chapter 58) should be performed to flush out unabsorbed pills. Deferoxamine, a potent iron-chelating compound, can be given intravenously to bind iron that has already been absorbed and to promote its excretion in urine and feces. Activated charcoal, a highly effective adsorbent for most toxins, does not bind iron and thus is ineffective. Appropriate supportive therapy for gastrointestinal bleeding, metabolic acidosis, and shock must also be provided.

B. Chronic Iron Toxicity

Chronic iron toxicity (iron overload), also known as hemochromatosis, results when excess iron is deposited in the heart, liver, pancreas, and other organs. It can lead to organ failure and death. It most commonly occurs in patients with inherited hemochromatosis, a disorder characterized by excessive iron absorption, and in patients who receive many red cell transfusions over a long period of time (eg, individuals with β-thalassemia).

Chronic iron overload in the absence of anemia is most efficiently treated by intermittent phlebotomy. One unit of blood can be removed every week or so until all of the excess iron is removed. Iron chelation therapy using parenteral deferoxamine or the oral iron chelator deferasirox (see Chapter 57) is less efficient as well as more complicated, expensive, and hazardous, but it may be the only option for iron overload that cannot be managed by phlebotomy, as is the case for many individuals with inherited and acquired causes of refractory anemia such as thalassemia major, sickle cell anemia, aplastic anemia, etc.


Vitamin B12 (cobalamin) serves as a cofactor for several essential biochemical reactions in humans. Deficiency of vitamin B12 leads to megaloblastic anemia (Table 33–2), gastrointestinal symptoms, and neurologic abnormalities. Although deficiency of vitamin B12 due to an inadequate supply in the diet is unusual, deficiency of B12 in adults—especially older adults—due to inadequate absorption of dietary vitamin B12 is a relatively common and easily treated disorder.


Vitamin B12 consists of a porphyrin-like ring with a central cobalt atom attached to a nucleotide. Various organic groups may be covalently bound to the cobalt atom, forming different cobalamins. Deoxyadenosylcobalamin and methylcobalamin are the active forms of the vitamin in humans. Cyanocobalamin and hydroxocobalamin (both available for therapeutic use) and other cobalamins found in food sources are converted to the active forms. The ultimate source of vitamin B12 is from microbial synthesis; the vitamin is not synthesized by animals or plants. The chief dietary source of vitamin B12 is microbially derived vitamin B12 in meat (especially liver), eggs, and dairy products. Vitamin B12 is sometimes called extrinsic factor to differentiate it from intrinsic factor, a protein secreted by the stomach that is required for gastrointestinal uptake of dietary vitamin B12.


The average American diet contains 5–30 mcg of vitamin B12 daily, 1–5 mcg of which is usually absorbed. The vitamin is avidly stored, primarily in the liver, with an average adult having a total vitamin B12storage pool of 3000–5000 mcg. Only trace amounts of vitamin B12 are normally lost in urine and stool. Because the normal daily requirements of vitamin B12 are only about 2 mcg, it would take about 5 years for all of the stored vitamin B12 to be exhausted and for megaloblastic anemia to develop if B12 absorption were stopped. Vitamin B12 is absorbed after it complexes with intrinsic factor, a glycoprotein secreted by the parietal cells of the gastric mucosa. Intrinsic factor combines with the vitamin B12 that is liberated from dietary sources in the stomach and duodenum, and the intrinsic factor-vitamin B12complex is subsequently absorbed in the distal ileum by a highly selective receptor-mediated transport system. Vitamin B12 deficiency in humans most often results from malabsorption of vitamin B12 due either to lack of intrinsic factor or to loss or malfunction of the absorptive mechanism in the distal ileum. Nutritional deficiency is rare but may be seen in strict vegetarians after many years without meat, eggs, or dairy products.

Once absorbed, vitamin B12 is transported to the various cells of the body bound to a family of specialized glycoproteins, trans-cobalamin I, II, and III. Excess vitamin B12 is stored in the liver.


Two essential enzymatic reactions in humans require vitamin B12 (Figure 33–2). In one, methylcobalamin serves as an intermediate in the transfer of a methyl group from N5-methyltetrahydrofolate to homocysteine, forming methionine (Figure 33–2AFigure 33–3section 1). Without vitamin B12, conversion of the major dietary and storage folate—N5-methyltetrahydrofolate—to tetrahydrofolate, the precursor of folate cofactors, cannot occur. As a result, vitamin B12 deficiency leads to deficiency of folate cofactors necessary for several biochemical reactions involving the transfer of one-carbon groups. In particular, the depletion of tetrahydrofolate prevents synthesis of adequate supplies of the deoxythymidylate (dTMP) and purines required for DNA synthesis in rapidly dividing cells, as shown in Figure 33–3section 2. The accumulation of folate as N5-methyltetrahydrofolate and the associated depletion of tetrahydrofolate cofactors in vitamin B12 deficiency have been referred to as the “methylfolate trap.” This is the biochemical step whereby vitamin B12 and folic acid metabolism are linked, and it explains why the megaloblastic anemia of vitamin B12 deficiency can be partially corrected by ingestion of large amounts of folic acid. Folic acid can be reduced to dihydrofolate by the enzyme dihydrofolate reductase (Figure 33–3section 3) and thereby serve as a source of the tetrahydrofolate required for synthesis of the purines and dTMP required for DNA synthesis.


FIGURE 33–2 Enzymatic reactions that use vitamin B12. See text for details.


FIGURE 33–3 Enzymatic reactions that use folates. Section 1 shows the vitamin B12-dependent reaction that allows most dietary folates to enter the tetrahydrofolate cofactor pool and becomes the “folate trap” in vitamin B12deficiency. Section 2 shows the deoxythymidine monophosphate (dTMP) cycle. Section 3 shows the pathway by which folic acid enters the tetrahydrofolate cofactor pool. Double arrows indicate pathways with more than one intermediate step. dUMP, deoxyuridine monophosphate.

Vitamin B12 deficiency causes the accumulation of homocysteine due to reduced formation of methylcobalamin, which is required for the conversion of homocysteine to methionine (Figure 33–3section 1). The increase in serum homocysteine can be used to help establish a diagnosis of vitamin B12 deficiency (Table 33–2). There is evidence from observational studies that elevated serum homocysteine increases the risk of atherosclerotic cardiovascular disease. However, randomized clinical trials have not shown a definitive reduction in cardiovascular events (myocardial infarction, stroke) in patients receiving vitamin supplementation that lowers serum homocysteine.

The other reaction that requires vitamin B12 is isomerization of methylmalonyl-CoA to succinyl-CoA by the enzyme methylmalonyl-CoA mutase (Figure 33–2B). In vitamin B12 deficiency, this conversion cannot take place and the substrate, methylmalonyl-CoA, as well as methylmalonic acid accumulate. The increase in serum and urine concentrations of methylmalonic acid can be used to support a diagnosis of vitamin B12 deficiency (Table 33–2). In the past, it was thought that abnormal accumulation of methylmalonyl-CoA causes the neurologic manifestations of vitamin B12 deficiency. However, newer evidence instead implicates the disruption of the methionine synthesis pathway as the cause of neurologic problems. Whatever the biochemical explanation for neurologic damage, the important point is that administration of folic acid in the setting of vitamin B12 deficiency will not prevent neurologic manifestations even though it will largely correct the anemia caused by the vitamin B12 deficiency.

Clinical Pharmacology

Vitamin B12 is used to treat or prevent deficiency. The most characteristic clinical manifestation of vitamin B12 deficiency is megaloblastic, macrocytic anemia (Table 33–2), often with associated mild or moderate leukopenia or thrombocytopenia (or both), and a characteristic hypercellular bone marrow with an accumulation of megaloblastic erythroid and other precursor cells. The neurologic syndrome associated with vitamin B12 deficiency usually begins with paresthesias in peripheral nerves and weakness and progresses to spasticity, ataxia, and other central nervous system dysfunctions. Correction of vitamin B12 deficiency arrests the progression of neurologic disease, but it may not fully reverse neurologic symptoms that have been present for several months. Although most patients with neurologic abnormalities caused by vitamin B12 deficiency have megaloblastic anemia when first seen, occasional patients have few if any hematologic abnormalities.

Once a diagnosis of megaloblastic anemia is made, it must be determined whether vitamin B12 or folic acid deficiency is the cause. (Other causes of megaloblastic anemia are very rare.) This can usually be accomplished by measuring serum levels of the vitamins. The Schilling test, which measures absorption and urinary excretion of radioactively labeled vitamin B12, can be used to further define the mechanism of vitamin B12 malabsorption when this is found to be the cause of the megaloblastic anemia.

The most common causes of vitamin B12 deficiency are pernicious anemia, partial or total gastrectomy, and conditions that affect the distal ileum, such as malabsorption syndromes, inflammatory bowel disease, or small bowel resection.

Pernicious anemia results from defective secretion of intrinsic factor by the gastric mucosal cells. Patients with pernicious anemia have gastric atrophy and fail to secrete intrinsic factor (as well as hydrochloric acid). These patients frequently have autoantibodies to intrinsic factor. The Schilling test shows diminished absorption of radioactively labeled vitamin B12, which is corrected when intrinsic factor is administered with radioactive B12, since the vitamin can then be normally absorbed.

Vitamin B12 deficiency also occurs when the region of the distal ileum that absorbs the vitamin B12-intrinsic factor complex is damaged, as when the ileum is involved with inflammatory bowel disease or when the ileum is surgically resected. In these situations, radioactively labeled vitamin B12 is not absorbed in the Schilling test, even when intrinsic factor is added. Rare cases of vitamin B12 deficiency in children have been found to be secondary to congenital deficiency of intrinsic factor or to defects of the receptor sites for vitamin B12-intrinsic factor complex located in the distal ileum. Because it is associated with use of radioactive isotopes, the Schilling test is unavailable in many centers. Alternatively one can test for intrinsic factor antibodies, and for elevated homocysteine and methylmalonic acid levels (Figure 33–2) to make a diagnosis of pernicious anemia with high sensitivity and specificity.

Almost all cases of vitamin B12 deficiency are caused by malabsorption of the vitamin; therefore, parenteral injections of vitamin B12 are required for therapy. For patients with potentially reversible diseases, the underlying disease should be treated after initial treatment with parenteral vitamin B12. Most patients, however, do not have curable deficiency syndromes and require lifelong treatment with vitamin B12.

Vitamin B12 for parenteral injection is available as cyanocobalamin or hydroxocobalamin. Hydroxocobalamin is preferred because it is more highly protein-bound and therefore remains longer in the circulation. Initial therapy should consist of 100–1000 mcg of vitamin B12 intramuscularly daily or every other day for 1–2 weeks to replenish body stores. Maintenance therapy consists of 100–1000 mcg intramuscularly once a month for life. If neurologic abnormalities are present, maintenance therapy injections should be given every 1–2 weeks for 6 months before switching to monthly injections. Oral vitamin B12-intrinsic factor mixtures and liver extracts should not be used to treat vitamin B12 deficiency; however, oral doses of 1000 mcg of vitamin B12 daily are usually sufficient to treat patients with pernicious anemia who refuse or cannot tolerate the injections. After pernicious anemia is in remission following parenteral vitamin B12 therapy, the vitamin can be administered intranasally as a spray or gel.


Reduced forms of folic acid are required for essential biochemical reactions that provide precursors for the synthesis of amino acids, purines, and DNA. Folate deficiency is relatively common, even though the deficiency is easily corrected by administration of folic acid. The consequences of folate deficiency go beyond the problem of anemia because folate deficiency is implicated as a cause of congenital malformations in newborns and may play a role in vascular disease (see Box: Folic Acid Supplementation: A Public Health Dilemma).


Folic acid (pteroylglutamic acid) is composed of a heterocycle (pteridine), p-aminobenzoic acid, and glutamic acid (Figure 33–4). Various numbers of glutamic acid moieties are attached to the pteroyl portion of the molecule, resulting in monoglutamates, triglutamates, or polyglutamates. Folic acid undergoes reduction, catalyzed by the enzyme dihydrofolate reductase (“folate reductase”), to give dihydrofolic acid (Figure 33–3section 3). Tetrahydrofolate is subsequently transformed to folate cofactors possessing one-carbon units attached to the 5-nitrogen, to the 10-nitrogen, or to both positions (Figure 33–3). Folate cofactors are interconvertible by various enzymatic reactions and serve the important biochemical function of donating one-carbon units at various levels of oxidation. In most of these, tetrahydrofolate is regenerated and becomes available for reutilization.


FIGURE 33–4 The structure of folic acid. (Reproduced, with permission, from Murray RK et al: Harper’s Biochemistry, 24th ed. McGraw-Hill, 1996. Copyright © The McGraw-Hill Companies, Inc.)

Folic Acid Supplementation: A Public Health Dilemma

Starting in January 1998, all products made from enriched grains in the United States and Canada were required to be supplemented with folic acid. These rulings were issued to reduce the incidence of congenital neural tube defects (NTDs). Epidemiologic studies show a strong correlation between maternal folic acid deficiency and the incidence of NTDs such as spina bifida and anencephaly. The requirement for folic acid supplementation is a public health measure aimed at the significant number of women who do not receive prenatal care and are not aware of the importance of adequate folic acid ingestion for preventing birth defects in their infants. Observational studies from countries that supplement grains with folic acid have found that supplementation is associated with a significant (20–25%) reduction in NTD rates. Observational studies also suggest that rates of other types of congenital anomalies (heart and orofacial) have fallen since supplementation began.

There may be an added benefit for adults. N5-Methyl-tetrahydrofolate is required for the conversion of homocysteine to methionine (Figure 33–2Figure 33–3, reaction 1). Impaired synthesis of N5-methyltetrahydrofolate results in elevated serum concentrations of homocysteine. Data from several sources suggest a positive correlation between elevated serum homocysteine and occlusive vascular diseases such as ischemic heart disease and stroke. Clinical data suggest that the folate supplementation program has improved the folate status and reduced the prevalence of hyperhomocysteinemia in a population of middle-aged and older adults who did not use vitamin supplements. There is also evidence that adequate folic acid protects against several cancers, including colorectal, breast, and cervical cancer.

Although the potential benefits of supplemental folic acid during pregnancy are compelling, the decision to require folic acid in grains was controversial. As described in the text, ingestion of folic acid can partially or totally correct the anemia caused by vitamin B12 deficiency. However, folic acid supplementation does not prevent the potentially irreversible neurologic damage caused by vitamin B12 deficiency. People with pernicious anemia and other forms of vitamin B12 deficiency are usually identified because of signs and symptoms of anemia, which typically occur before neurologic symptoms. Some opponents of folic acid supplementation were concerned that increased folic acid intake in the general population would mask vitamin B12 deficiency and increase the prevalence of neurologic disease in the elderly population. To put this in perspective, approximately 4000 pregnancies, including 2500 live births, in the United States each year are affected by NTDs. In contrast, it is estimated that over 10% of the elderly population in the United States, or several million people, are at risk for the neuropsychiatric complications of vitamin B12 deficiency. In acknowledgment of this controversy, the FDA kept its requirements for folic acid supplementation at a somewhat low level. There is also concern based on observational and prospective clinical trials that high folic acid levels can increase the risk of some diseases, such as colorectal cancer, for which folic acid may exhibit a bell-shaped curve. Further research is needed to more accurately define the optimal level of folic acid fortification in food and recommendations for folic acid supplementation in different populations and age groups.


The average American diet contains 500–700 mcg of folates daily, 50–200 mcg of which is usually absorbed, depending on metabolic requirements. Pregnant women may absorb as much as 300–400 mcg of folic acid daily. Various forms of folic acid are present in a wide variety of plant and animal tissues; the richest sources are yeast, liver, kidney, and green vegetables. Normally, 5–20 mg of folates is stored in the liver and other tissues. Folates are excreted in the urine and stool and are also destroyed by catabolism, so serum levels fall within a few days when intake is diminished. Because body stores of folates are relatively low and daily requirements high, folic acid deficiency and megaloblastic anemia can develop within 1–6 months after the intake of folic acid stops, depending on the patient’s nutritional status and the rate of folate utilization.

Unaltered folic acid is readily and completely absorbed in the proximal jejunum. Dietary folates, however, consist primarily of polyglutamate forms of N5-methyltetrahydrofolate. Before absorption, all but one of the glutamyl residues of the polyglutamates must be hydrolyzed by the enzyme α-1-glutamyl transferase (“conjugase”) within the brush border of the intestinal mucosa. The monoglutamate N5-methyltetrahydrofolate is subsequently transported into the bloodstream by both active and passive transport and is then widely distributed throughout the body. Inside cells, N5-methyltetrahydro-folate is converted to tetrahydrofolate by the demethylation reaction that requires vitamin B12 (Figure 33–3section 1).


Tetrahydrofolate cofactors participate in one-carbon transfer reactions. As described earlier in the discussion of vitamin B12, one of these essential reactions produces the dTMP needed for DNA synthesis. In this reaction, the enzyme thymidylate synthase catalyzes the transfer of the one-carbon unit of N5N10-methylenetetrahydrofolate to deoxyuridine monophosphate (dUMP) to form dTMP (Figure 33–3section 2). Unlike all the other enzymatic reactions that use folate cofactors, in this reaction the cofactor is oxidized to dihydrofolate, and for each mole of dTMP produced, 1 mole of tetrahydrofolate is consumed. In rapidly proliferating tissues, considerable amounts of tetrahydrofolate are consumed in this reaction, and continued DNA synthesis requires continued regeneration of tetrahydrofolate by reduction of dihydrofolate, catalyzed by the enzyme dihydrofolate reductase. The tetrahydrofolate thus produced can then reform the cofactor N5N10-methylenetetrahydrofolate by the action of serine transhydroxymethylase and thus allow for the continued synthesis of dTMP. The combined catalytic activities of dTMP synthase, dihydrofolate reductase, and serine transhydroxymethylase are referred to as the dTMP synthesis cycle. Enzymes in the dTMP cycle are the targets of two anti-cancer drugs; methotrexate inhibits dihydrofolate reductase, and a metabolite of 5-fluorouracil inhibits thymidylate synthase (see Chapter 54).

Cofactors of tetrahydrofolate participate in several other essential reactions. N5-Methylenetetrahydrofolate is required for the vitamin B12-dependent reaction that generates methionine from homocysteine (Figure 33–2AFigure 33–3section 1). In addition, tetrahydrofolate cofactors donate one-carbon units during the de novo synthesis of essential purines. In these reactions, tetrahydrofolate is regenerated and can reenter the tetrahydrofolate cofactor pool.

Clinical Pharmacology

Folate deficiency results in a megaloblastic anemia that is microscopically indistinguishable from the anemia caused by vitamin B12 deficiency (see above). However, folate deficiency does not cause the characteristic neurologic syndrome seen in vitamin B12 deficiency. In patients with megaloblastic anemia, folate status is assessed with assays for serum folate or for red blood cell folate. Red blood cell folate levels are often of greater diagnostic value than serum levels, because serum folate levels tend to be labile and do not necessarily reflect tissue levels.

Folic acid deficiency is often caused by inadequate dietary intake of folates. Patients with alcohol dependence and patients with liver disease can develop folic acid deficiency because of poor diet and diminished hepatic storage of folates. Pregnant women and patients with hemolytic anemia have increased folate requirements and may become folic acid-deficient, especially if their diets are marginal. Evidence implicates maternal folic acid deficiency in the occurrence of fetal neural tube defects. (See Box: Folic Acid Supplementation: A Public Health Dilemma.) Patients with malabsorption syndromes also frequently develop folic acid deficiency. Patients who require renal dialysis are at risk of folic acid deficiency because folates are removed from the plasma during the dialysis procedure.

Folic acid deficiency can be caused by drugs. Methotrexate and, to a lesser extent, trimethoprim and pyrimethamine, inhibit dihydrofolate reductase and may result in a deficiency of folate cofactors and ultimately in megaloblastic anemia. Long-term therapy with phenytoin can also cause folate deficiency, but only rarely causes megaloblastic anemia.

Parenteral administration of folic acid is rarely necessary, since oral folic acid is well absorbed even in patients with malabsorption syndromes. A dose of 1 mg folic acid orally daily is sufficient to reverse megaloblastic anemia, restore normal serum folate levels, and replenish body stores of folates in almost all patients. Therapy should be continued until the underlying cause of the deficiency is removed or corrected. Therapy may be required indefinitely for patients with malabsorption or dietary inadequacy. Folic acid supplementation to prevent folic acid deficiency should be considered in high-risk patients, including pregnant women, patients with alcohol dependence, hemolytic anemia, liver disease, or certain skin diseases, and patients on renal dialysis.


The hematopoietic growth factors are glycoprotein hormones that regulate the proliferation and differentiation of hematopoietic progenitor cells in the bone marrow. The first growth factors to be identified were called colony-stimulating factors because they could stimulate the growth of colonies of various bone marrow progenitor cells in vitro. Many of these growth factors have been purified and cloned, and their effects on hematopoiesis have been extensively studied. Quantities of these growth factors sufficient for clinical use are produced by recombinant DNA technology.

Of the known hematopoietic growth factors, erythropoietin (epoetin alfa and epoetin beta), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin-11 (IL-11) and thrombopoietin receptor agonists (romiplostim and eltrombopag) are currently in clinical use.

The hematopoietic growth factors and drugs that mimic their action have complex effects on the function of a wide variety of cell types, including nonhematologic cells. Their usefulness in other areas of medicine, particularly as potential anti-cancer and anti-inflammatory drugs, is being investigated.


Chemistry & Pharmacokinetics

Erythropoietin, a 34–39 kDa glycoprotein, was the first human hematopoietic growth factor to be isolated. It was originally purified from the urine of patients with severe anemia. Recombinant human erythropoietin (rHuEPO, epoetin alfa) is produced in a mammalian cell expression system. After intravenous administration, erythropoietin has a serum half-life of 4–13 hours in patients with chronic renal failure. It is not cleared by dialysis. It is measured in international units (IU). Darbepoetin alfa is a modified form of erythropoietin that is more heavily glycosylated as a result of changes in amino acids. Darbepoetin alfa has a twofold to threefold longer half-life than epoetin alfa. Methoxy polyethylene glycol-epoetin beta is an isoform of erythropoietin covalently attached to a long polyethylene glycol polymer. This long-lived recombinant product is administered as a single intravenous or subcutaneous dose at 2-week or monthly intervals, whereas epoetin alfa is generally administered three times a week and darbepoetin is administered weekly.


Erythropoietin stimulates erythroid proliferation and differentiation by interacting with erythropoietin receptors on red cell progenitors. The erythropoietin receptor is a member of the JAK/STAT superfamily of cytokine receptors that use protein phosphorylation and transcription factor activation to regulate cellular function (see Chapter 2). Erythropoietin also induces release of reticulocytes from the bone marrow. Endogenous erythropoietin is primarily produced in the kidney. In response to tissue hypoxia, more erythropoietin is produced through an increased rate of transcription of the erythropoietin gene. This results in correction of the anemia, provided that the bone marrow response is not impaired by red cell nutritional deficiency (especially iron deficiency), primary bone marrow disorders (see below), or bone marrow suppression from drugs or chronic diseases.

Normally, an inverse relationship exists between the hematocrit or hemoglobin level and the serum erythropoietin level. Nonanemic individuals have serum erythropoietin levels of less than 20 IU/L. As the hematocrit and hemoglobin levels fall and anemia becomes more severe, the serum erythropoietin level rises exponentially. Patients with moderately severe anemia usually have erythropoietin levels in the 100–500 IU/L range, and patients with severe anemia may have levels of thousands of IU/L. The most important exception to this inverse relationship is in the anemia of chronic renal failure. In patients with renal disease, erythropoietin levels are usually low because the kidneys cannot produce the growth factor. These are the patients most likely to respond to treatment with exogenous erythropoietin. In most primary bone marrow disorders (aplastic anemia, leukemias, myeloproliferative and myelodysplastic disorders, etc) and most nutritional and secondary anemias, endogenous erythropoietin levels are high, so there is less likelihood of a response to exogenous erythropoietin (but see below).

Clinical Pharmacology

The availability of erythropoiesis-stimulating agents (ESAs) has had a significant positive impact for patients with several types of anemia (Table 33–4). The ESAs consistently improve the hematocrit and hemoglobin level, often eliminate the need for transfusions, and reliably improve quality of life indices. The ESAs are used routinely in patients with anemia secondary to chronic kidney disease. In patients treated with an ESA, an increase in reticulocyte count is usually observed in about 10 days and an increase in hematocrit and hemoglobin levels in 2–6 weeks. Dosages of ESAs are adjusted to maintain a target hemoglobin up to, but not exceeding, 10–12 g/dL. To support the increased erythropoiesis, nearly all patients with chronic kidney disease require oral or parenteral iron supplementation. Folate supplementation may also be necessary in some patients.

TABLE 33–4 Clinical uses of hematopoietic growth factors and agents that mimic their actions.


In selected patients, erythropoietin is also used to reduce the need for red blood cell transfusion in patients undergoing myelosuppressive cancer chemotherapy who have a hemoglobin level of less than 10 g/dL, and for selected patients with low-risk myelodysplastic syndromes and anemia requiring red blood cell transfusion. Patients who have disproportionately low serum erythropoietin levels for their degree of anemia are most likely to respond to treatment. Patients with endogenous erythropoietin levels of less than 100 IU/L have the best chance of response, although patients with erythropoietin levels between 100 and 500 IU/L respond occasionally. Methoxy polyethylene glycol-epoetin beta should not be used for treatment of anemia caused by cancer chemotherapy because a clinical trial found significantly more deaths among patients receiving this form of erythropoietin.

Erythropoietin is one of the drugs commonly used illegally by endurance athletes to enhance performance. Other methods such as autologous transfusion of red cells or use of androgens have also been used to increase hemoglobin. “Blood doping” constitutes a serious health risk to athletes and as a form of cheating is universally banned and routinely tested for in athletic events.


The most common adverse effects of erythropoietin are hypertension and thrombotic complications. ESAs increase the risk of serious cardiovascular events, thromboembolic events, stroke, and mortality in clinical studies when given to support hemoglobin levels greater than 11 g/dL In addition, a meta-analysis of 51 placebo-controlled trials of ESAs in cancer patients reported an increased rate of all-cause mortality and venous thrombosis in those receiving an ESA. Based on the accumulated evidence, it is recommended that the hemoglobin level not exceed 11 g/dL in patients with chronic kidney disease receiving an ESA, and that ESAs be used conservatively in cancer patients (eg, when hemoglobin levels are < 10 g/dL) and with the lowest dose needed to avoid transfusion. It is further recommended that ESAs not be used when a cancer therapy is being given with curative intent.

Allergic reactions to ESAs have been infrequent. There have been a small number of cases of pure red cell aplasia (PRCA) accompanied by neutralizing antibodies to erythropoietin. PRCA was most commonly seen in dialysis patients treated subcutaneously for a long period with a particular form of epoetin alfa (Eprex with a polysorbate 80 stabilizer rather than human serum albumin) that is not available in the United States. After regulatory agencies required that Eprex be administered intravenously rather than subcutaneously, the rate of ESA-associated PRCA diminished. However, rare cases have still been seen with all ESAs administered subcutaneously for long periods to patients with chronic kidney disease.


Chemistry & Pharmacokinetics

G-CSF and GM-CSF, the two myeloid growth factors currently available for clinical use, were originally purified from cultured human cell lines (Table 33–4). Recombinant human G-CSF (rHuG-CSF; filgrastim) is produced in a bacterial expression system. It is a nonglycosylated peptide of 175 amino acids, with a molecular weight of 18 kDa. Recombinant human GM-CSF (rHuGM-CSF; sargramostim) is produced in a yeast expression system. It is a partially glycosylated peptide of 127 amino acids, comprising three molecular species with molecular weights of 15,500, 15,800, and 19,500. These preparations have serum half-lives of 2–7 hours after intravenous or subcutaneous administration. Pegfilgrastim, a covalent conjugation product of filgrastim and a form of polyethylene glycol, has a much longer serum half-life than recombinant G-CSF, and it can be injected once per myelosuppressive chemotherapy cycle instead of daily for several days. Lenograstim, used widely in Europe, is a glycosylated form of recombinant G-CSF.


The myeloid growth factors stimulate proliferation and differentiation by interacting with specific receptors found on myeloid progenitor cells. Like the erythropoietin receptor, these receptors are members of the JAK/STAT superfamily (see Chapter 2). G-CSF stimulates proliferation and differentiation of progenitors already committed to the neutrophil lineage. It also activates the phagocytic activity of mature neutrophils and prolongs their survival in the circulation. G-CSF also has a remarkable ability to mobilize hematopoietic stem cells, ie, to increase their concentration in peripheral blood. This biologic effect underlies a major advance in transplantation—the use of peripheral blood stem cells (PBSCs) rather than bone marrow stem cells for autologous and allogeneic hematopoietic stem cell transplantation (see below).

GM-CSF has broader biologic actions than G-CSF. It is a multipotential hematopoietic growth factor that stimulates proliferation and differentiation of early and late granulocytic progenitor cells as well as erythroid and megakaryocyte progenitors. Like G-CSF, GM-CSF also stimulates the function of mature neutrophils. GM-CSF acts together with interleukin-2 to stimulate T-cell proliferation and appears to be a locally active factor at the site of inflammation. GM-CSF mobilizes peripheral blood stem cells, but it is significantly less efficacious and more toxic than G-CSF in this regard.

Clinical Pharmacology

A. Cancer Chemotherapy-Induced Neutropenia

Neutropenia is a common adverse effect of the cytotoxic drugs used to treat cancer and increases the risk of serious infection in patients receiving chemotherapy. Unlike the treatment of anemia and thrombocytopenia, transfusion of neutropenic patients with granulocytes collected from donors is performed rarely and with limited success. The introduction of G-CSF in 1991 represented a milestone in the treatment of chemotherapy-induced neutropenia. This growth factor dramatically accelerates the rate of neutrophil recovery after dose-intensive myelosuppressive chemotherapy (Figure 33–5). It reduces the duration of neutropenia and usually raises the nadir count, the lowest neutrophil count seen following a cycle of chemotherapy.


FIGURE 33–5 Effects of granulocyte colony-stimulating factor (G-CSF; red line) or placebo (green line) on absolute neutrophil count (ANC) after cytotoxic chemotherapy for lung cancer. Doses of chemotherapeutic drugs were administered on days 1 and 3. G-CSF or placebo injections were started on day 4 and continued daily through day 12 or 16. The first peak in ANC reflects the recruitment of mature cells by G-CSF. The second peak reflects a marked increase in new neutrophil production by the bone marrow under stimulation by G-CSF. (Normal ANC is 2.2–8.6 × 109/L.) (Reproduced, with permission, from Crawford J et al: Reduction by granulocyte colony-stimulating factor of fever and neutropenia induced by chemotherapy in patients with small-cell lung cancer. N Engl J Med 1991;325:164. Copyright © 1991 Massachusetts Medical Society. Reprinted with permission from Massachusetts Medical Society.)

The ability of G-CSF to increase neutrophil counts after myelosuppressive chemotherapy is nearly universal, but its impact on clinical outcomes is more variable. Many, but not all, clinical trials and meta-analyses have shown that G-CSF reduces episodes of febrile neutropenia, requirements for broad-spectrum antibiotics, infections, and days of hospitalization. Clinical trials have not shown improved survival in cancer patients treated with G-CSF. Clinical guidelines for the use of G-CSF after cytotoxic chemotherapy recommend reserving G-CSF for patients at high risk for febrile neutropenia based on age, medical history, and disease characteristics; patients receiving dose-intensive chemotherapy regimens that carry a greater than 20% risk of causing febrile neutropenia; patients with a prior episode of febrile neutropenia after cytotoxic chemotherapy; patients at high risk for febrile neutropenia; and patients who are unlikely to survive an episode of febrile neutropenia. Pegfilgrastim is an alternative to G-CSF for prevention of chemotherapy-induced febrile neutropenia. Pegfilgrastim can be administered once per chemotherapy cycle, and it may shorten the period of severe neutropenia slightly more than G-CSF.

Like G-CSF and pegfilgrastim, GM-CSF also reduces the duration of neutropenia after cytotoxic chemotherapy. It has been more difficult to show that GM-CSF reduces the incidence of febrile neutropenia, probably because GM-CSF itself can induce fever. In the treatment of chemotherapy-induced neutropenia, G-CSF, 5 mcg/kg/d, or GM-CSF, 250 mcg/m2/d, is usually started within 24–72 hours after completing chemotherapy and is continued until the absolute neutrophil count is greater than 10,000 cells/μL. Pegfilgrastim is given as a single dose of 6 mg.

The utility and safety of the myeloid growth factors in the postchemotherapy supportive care of patients with acute myeloid leukemia (AML) have been the subject of a number of clinical trials. Because leukemic cells arise from progenitors whose proliferation and differentiation are normally regulated by hematopoietic growth factors, including GM-CSF and G-CSF, there was concern that myeloid growth factors could stimulate leukemic cell growth and increase the rate of relapse. The results of randomized clinical trials suggest that both G-CSF and GM-CSF are safe following induction and consolidation treatment of myeloid and lymphoblastic leukemia. There has been no evidence that these growth factors reduce the rate of remission or increase relapse rate. On the contrary, the growth factors accelerate neutrophil recovery and reduce infection rates and days of hospitalization. Both G-CSF and GM-CSF have FDA approval for treatment of patients with AML.

B. Other Applications

G-CSF and GM-CSF have also proved to be effective in treating the neutropenia associated with congenital neutropeniacyclic neutropeniamyelodysplasia, and aplastic anemia. Many patients with these disorders respond with a prompt and sometimes dramatic increase in neutrophil count. In some cases, this results in a decrease in the frequency of infections. Because neither G-CSF nor GM-CSF stimulates the formation of erythrocytes and platelets, they are sometimes combined with other growth factors for treatment of pancytopenia.

The myeloid growth factors play an important role in autologous stem cell transplantation for patients undergoing high-dose chemotherapy. High-dose chemotherapy with autologous stem cell support is increasingly used to treat patients with tumors that are resistant to standard doses of chemotherapeutic drugs. The high-dose regimens produce extreme myelosuppression; the myelosuppression is then counteracted by reinfusion of the patient’s own hematopoietic stem cells (which are collected prior to chemotherapy). The administration of G-CSF or GM-CSF early after autologous stem cell transplantation reduces the time to engraftment and to recovery from neutropenia in patients receiving stem cells obtained either from bone marrow or from peripheral blood. These effects are seen in patients being treated for lymphoma or for solid tumors. G-CSF and GM-CSF are also used to support patients who have received allogeneic bone marrow transplantation for treatment of hematologic malignancies or bone marrow failure states. In this setting, the growth factors speed the recovery from neutropenia without increasing the incidence of acute graft-versus-host disease.

Perhaps the most important role of the myeloid growth factors in transplantation is for mobilization of PBSCs. Stem cells collected from peripheral blood have nearly replaced bone marrow as the hematopoietic preparation used for autologous and allogeneic transplantation. The cells can be collected in an outpatient setting with a procedure that avoids much of the risk and discomfort of bone marrow collection, including the need for general anesthesia. In addition, there is evidence that PBSC transplantation results in more rapid engraftment of all hematopoietic cell lineages and in reduced rates of graft failure or delayed platelet recovery.

G-CSF is the cytokine most commonly used for PBSC mobilization because of its increased efficacy and reduced toxicity compared with GM-CSF. To mobilize stem cells for autologous transplantation, donors are given 5–10 mcg/kg/d subcutaneously for 4 days. On the fifth day, they undergo leukapheresis. The success of PBSC transplantation depends on transfusion of adequate numbers of stem cells. CD34, an antigen present on early progenitor cells and absent from later, committed, cells, is used as a marker for the requisite stem cells. The goal is to infuse at least 5 × 106 CD34 cells/kg; this number of CD34 cells usually results in prompt and durable engraftment of all cell lineages. It may take several separate leukaphereses to collect enough CD34 cells, especially from older patients and patients who have been exposed to radiation therapy or chemotherapy.

For patients with multiple myeloma or non-Hodgkin’s lymphoma who respond suboptimally to G-CSF alone, the novel hematopoietic stem cell mobilizer plerixafor can be added to G-CSF. Plerixafor is a bicyclam molecule originally developed as an anti-HIV drug because of its ability to inhibit the CXC chemokine receptor 4 (CXCR4), a co-receptor for HIV entry into CD4+ T lymphocytes (see Chapter 49). Early clinical trials of plerixafor revealed a remarkable ability to increase CD34 cells in peripheral blood. Plerixafor mobilizes CD34 cells by preventing chemokine stromal cell-derived factor-1α (SDF-1α) from binding to CXCR4 and directing the CD34 cells to “home” to the bone marrow. Plerixafor is administered by subcutaneous injection after 4 days of G-CSF treatment and 11 hours prior to leukapheresis; it can be used with G-CSF for up to 4 continuous days. Plerixafor is eliminated primarily by the renal route and must be dose-adjusted for patients with renal impairment. The drug is well-tolerated; the most common adverse effects associated with its use are injection site reactions, gastrointestinal disturbances, dizziness, fatigue, and headache.


Although the three growth factors have similar effects on neutrophil counts, G-CSF and pegfilgrastim are used more frequently than GM-CSF because they are better tolerated. G-CSF and pegfilgrastim can cause bone pain, which clears when the drugs are discontinued. GM-CSF can cause more severe side effects, particularly at higher doses. These include fever, malaise, arthralgias, myalgias, and a capillary leak syndrome characterized by peripheral edema and pleural or pericardial effusions. Allergic reactions may occur but are infrequent. Splenic rupture is a rare but serious complication of the use of G-CSF for PBSC.


Patients with thrombocytopenia have a high risk of hemorrhage. Although platelet transfusion is commonly used to treat thrombocytopenia, this procedure can cause adverse reactions in the recipient; furthermore, a significant number of patients fail to exhibit the expected increase in platelet count. Thrombopoietin (TPO) and IL-11 both appear to be key endogenous regulators of platelet production. A recombinant form of IL-11 was the first agent to gain FDA approval for treatment of thrombocytopenia. Recombinant human thrombopoietin and a pegylated form of a shortened human thrombopoietin protein underwent extensive clinical investigation in the 1990s. However, further development was abandoned after autoantibodies to the native thrombopoietin formed in healthy human subjects and caused thrombocytopenia. Efforts shifted to investigation of novel, nonimmunogenic agonists of the thrombopoietin receptor, which is known as Mpl. Two thrombopoietin agonists (romiplostim and eltrombopag) are approved for treatment of thrombocytopenia.

Chemistry & Pharmacokinetics

Interleukin-11 is a 65–85 kDa protein produced by fibroblasts and stromal cells in the bone marrow. Oprelvekin, the recombinant form of IL-11 approved for clinical use (Table 33–4), is produced by expression in Escherichia coli. The half-life of IL-11 is 7–8 hours when the drug is injected subcutaneously.

Romiplostim is a peptide covalently linked to antibody fragments that serve to extend the peptide’s half-life. The Mpl-binding peptide has no sequence homology with human thrombopoietin and there is no evidence in animal or human studies that the Mpl-binding peptide or romiplostim induces antibodies to thrombopoietin. After subcutaneous administration, romiplostim is eliminated by the reticuloendothelial system with an average half-life of 3–4 days. Its half-life is inversely related to the serum platelet count; it has a longer half-life in patients with thrombocytopenia and a shorter half-life in patients whose platelet counts have recovered to normal levels. Romiplostim is approved for therapy of patients with chronic immune thrombocytopenia who have had an inadequate response to other therapies.

Eltrombopag is an orally active small nonpeptide thrombopoietin agonist molecule approved for therapy of patients with chronic immune thrombocytopenia who have had an inadequate response to other therapies, and for treatment of thrombocytopenia in patients with hepatitis C to allow initiation of interferon therapy. Following oral administration, peak eltrombopag levels are observed in 2–6 hours and half-life is 26–35 hours. Eltrombopag is primarily excreted in the feces.


Interleukin-11 acts through a specific cell surface cytokine receptor to stimulate the growth of multiple lymphoid and myeloid cells. It acts synergistically with other growth factors to stimulate the growth of primitive megakaryocytic progenitors and, most importantly, increases the number of peripheral platelets and neutrophils.

Romiplostim has high affinity for the human Mpl receptor. Eltrombopag interacts with the transmembrane domain of the Mpl receptor. Both drugs induce signaling through the Mpl receptor pathway and cause a dose-dependent increase in platelet count. Romiplostim is administered once weekly by subcutaneous injection. Eltrombopag is an oral drug. For both drugs, peak platelet count responses are observed in approximately 2 weeks.

Clinical Pharmacology

Interleukin-11 is approved for the secondary prevention of thrombocytopenia in patients receiving cytotoxic chemotherapy for treatment of nonmyeloid cancers. Clinical trials show that it reduces the number of platelet transfusions required by patients who experience severe thrombocytopenia after a previous cycle of chemotherapy. Although IL-11 has broad stimulatory effects on hematopoietic cell lineages in vitro, it does not appear to have significant effects on the leukopenia caused by myelosuppressive chemotherapy. Interleukin-11 is given by subcutaneous injection at a dose of 50 mcg/kg/d. It is started 6–24 hours after completion of chemotherapy and continued for 14–21 days or until the platelet count passes the nadir and rises to more than 50,000 cells/μL.

In patients with chronic immune thrombocytopenia who failed to respond adequately to previous treatment with steroids, immunoglobulins, or splenectomy, romiplostim and eltrombopag significantly increase platelet count in most patients. Both drugs are used at the minimal dose required to maintain platelet counts of greater than 50,000 cells/μL.


The most common adverse effects of IL-11 are fatigue, headache, dizziness, and cardiovascular effects. The cardiovascular effects include anemia (due to hemodilution), dyspnea (due to fluid accumulation in the lungs), and transient atrial arrhythmias. Hypokalemia has also been seen in some patients. All of these adverse effects appear to be reversible.

Eltrombopag is potentially hepatotoxic and liver function must be monitored, particularly when used in patients with hepatitis C. Portal vein thrombosis has also been reported with eltrombopag and romiplostim in the setting of chronic liver disease. In patients with myelodysplastic syndromes, romiplostim increases the blast count and risk of progression to acute myeloid leukemia. Marrow fibrosis has been observed with thrombopoietin agonists but is generally reversible when the drug is discontinued. Rebound thrombocytopenia has been observed following discontinuation of TPO agonists.

SUMMARY Agents Used in Anemias and Hematopoietic Growth Factors







Aapro MS et al, European Organisation for Research and Treatment of Cancer: 2010 update of EORTC guidelines for the use of granulocyte-colony stimulating factor to reduce the incidence of chemotherapy-induced febrile neutropenia in adult patients with lymphoproliferative disorders and solid tumours. Eur J Cancer 2011;47:8.

Albaramki J et al: Parenteral versus oral iron therapy for adults and children with chronic kidney disease. Cochrane Database Syst Rev 2012;(1):CD007857.

Auerbach M, Al Talib K: Low-molecular weight iron dextran and iron sucrose have similar comparative safety profiles in chronic kidney disease. Kidney Int 2008;73:528.

Barzi A, Sekeres MA: Myelodysplastic syndromes: A practical approach to diagnosis and treatment. Cleve Clin J Med 2010;77:37.

Brittenham GM: Iron-chelating therapy for transfusional iron overload. N Engl J Med 2011;364:146.

Clark SF: Iron deficiency anemia: diagnosis and management. Curr Opin Gastroenterol 2009;25:122.

Darshan D, Fraer DM, Anderson GJ: Molecular basis of iron-loading disorders. Expert Rev Mol Med 2010;12:e36.

Gertz MA: Current status of stem cell mobilization. Br J Haematol 2010;150:647.

Kessans MR, Gatesman ML, Kockler DR: Plerixafor: A peripheral blood stem cell mobilizer. Pharmacotherapy 2010;30:485.

McKoy JM et al: Epoetin-associated pure red cell aplasia: Past, present, and future considerations. Transfusion 2008;48:1754.

Rees DC, Williams TN, Gladwin MT: Sickle-cell disease. Lancet 2010;376:2018.

Rizzo JD et al: American Society of Clinical Oncology/American Society of Hematology clinical practice guideline update on the use of epoetin and darbepoetin in adult patients with cancer. J Clin Oncol 2010;28:4996.

Sauer J, Mason JB, Choi SW: Too much folate: A risk factor for cancer and cardiovascular disease? Curr Opin Clin Nutr Metab Care 2009;12:30.

Solomon LR: Disorders of cobalamin (vitamin B12) metabolism: Emerging concepts in pathophysiology, diagnosis and treatment. Blood Rev 2007;21:113.

Stasi R et al: Thrombopoietic agents. Blood Rev 2010;24:179.

Wolff T et al: Folic acid supplementation for the prevention of neural tube defects: An update of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med 2009;150:632.


This patient’s megaloblastic anemia appears to be due to vitamin B12 (cobalamin) deficiency secondary to impaired production of intrinsic factor, resulting in insufficient absorption of vitamin B12 from the gastrointestinal tract. It is important to measure serum concentrations of both folic acid and cobalamin because megaloblastic anemia can result from deficiency of either nutrient. It is especially important to diagnose vitamin B12 deficiency because this deficiency, if untreated, can lead to irreversible neurologic damage. Folate supplementation, which can compensate for vitamin B12-derived anemia, does not prevent B12-deficiency neurologic damage. To correct this patient’s vitamin B12 deficiency, she would probably be treated parenterally with cobalamin because of her impaired oral absorption of vitamin B12. Several weeks of daily administration would be followed with weekly doses until her hematocrit returned to normal. Monthly doses would then be given to maintain her body stores of vitamin B12.


*The author acknowledges contributions of the previous author of this chapter, Susan B. Masters, PhD.