Abeloff's Clinical Oncology, 4th Edition

Part II – Problems Common to Cancer and its Therapy

Section G – Complications of Therapy

Chapter 66 – Second Malignant Neoplasms

John P. Plastaras,Daniel M. Green,
Giulio J. D'Angio

SUMMARY OF KEY POINTS

Epidemiology

  

   

Pediatric patients who survive their primary cancer are at increased risk of developing a new malignancy.

  

   

The magnitude of this risk is modulated by several factors, including the patient's genetic susceptibility, the type of surgical procedure used for removal of the tumor, the use of radiation therapy as part of the treatment plan, the chemotherapeutic agents that are employed, the severity of immune suppression that is present at the completion of all treatment, and environmental exposures.

  

   

The potential risk of a second malignant process following treatment of adults should not lead to therapeutic compromises when treatment, often aggressive, is of known benefit.

General Recommendations

  

   

All former cancer patients should remain under a physician's care indefinitely.

  

   

All former cancer patients should undergo frequent physical examinations (preferably by a physician or other specifically trained health care worker who is familiar with the problem of therapy-induced malignancy).

  

   

Physicians should counsel cancer survivors regarding the potential for the adverse effects of tobacco use, including the potential to interact with the adverse effects of their prior therapy, such as irradiation of the oropharynx, esophagus, and/or lungs.

Specific Recommendations

  

   

Plain radiographs should be obtained for patients who have been irradiated whenever local pain occurs in a previously irradiated bone.

  

   

Patients who have received irradiation to volumes that include the breast, uterine cervix, or intestine should undergo routine evaluation with available screening tests, such as mammography, Pap smear, and stool examination for the presence of occult blood.

  

   

Annual mammography should be initiated no later than 10 years after breast irradiation.

  

   

Careful physical examination of irradiated patients will facilitate the early identification of thyroid nodules and skin cancer.

  

   

All patients who have been treated with an alkylating agent, procarbazine, or a topoisomerase II inhibitor, should have a complete blood count every 6 to 12 months for a minimum of 12 years after diagnosis.

  

   

Presence of macrocytosis and/or cytopenia should prompt evaluation of the bone marrow.

INTRODUCTION

Second malignant neoplasms (SMNs) develop in patients as a result of genetic and iatrogenic factors and their interplay. The therapies that are employed are different for children and adults because of the differences in the primary cancers that are encountered. Primary cancers in adults are usually of epithelial origin, unlike the embryonal and sarcomatous neoplasms encountered at earlier ages. In addition, some of the cancers that are encountered in adults are the result of therapies they received for nonmalignant conditions that were treated when they were children. This discussion therefore has been divided into two parts according to the age group being considered.

CHILDREN AND ADOLESCENTS

SMNs are a recognized complication of successful treatment of children and adolescents for cancer. The frequency of these had been reported to be 1.9% at 10 years after diagnosis, 5.0% to 12.0% at 25 years after diagnosis, 3.3% to 4.9% at 25 years after diagnosis among 3-year survivors, 3.2% to 12.0% at 20 years after diagnosis among 5-year survivors, and 4.2% to 7.8% at 25 years after diagnosis among 5-year survivors. [1] [2] [3] [4] [5] [6] [7] [8] These series differed with respect to the time period during which the patients were treated, the completeness of follow-up, and the treatment exposures that the patients experienced. SMNs are a major threat to the long-term survivors of childhood cancer, in whom they are looming ever larger as a cause of death. Moller and colleagues reported results of a survey in the Nordic countries, where the proportions of deaths from second tumors increased from 3% to 22% between 5 and 10 years and 20+ years.[9] In fact, Lawless and colleagues found SMNs to be more often the cause of death than was relapse of the primary tumor.[10] In their single-institution study, they found that SMNs were the leading cause of death (39%) in 15+-year survivors; greater than the rates for primary cancer (21%) and cardiac (16%) deaths. Increasingly, the data suggest that although specific exposures (whether to a particular chemotherapeutic agent or to ionizing radiation) might be linked to the occurrence of a new malignancy, the most important factor in the pathogenesis of many SMNs could well be the patient's genetic susceptibility. We will review the genetic and treatment factors that have been associated with the occurrence of SMNs and discuss the follow-up and evaluation of the successfully treated pediatric cancer patient.

Genetic Factors

The importance of genetic predisposition to the occurrence of a SMN has been demonstrated most clearly in patients with hereditary retinoblastoma. Among 1604 1-year survivors of retinoblastoma in Boston and New York City between 1914 and 1984, 961 had the hereditary form of the disease.[11] The cumulative percentage of those who developed a SMN was 51.0% (±6.2%) 50 years after retinoblastoma diagnosis, compared with 5.0% (±3.0%) among those with nonhereditary retinoblastoma ( Fig. 66-1A ). Among patients with hereditary retinoblastoma, the cumulative percentage that developed SMNs was 58% among those whose treatment included radiation therapy (RT), compared with 26.5% among those whose treatment did not include RT ( Fig. 66-1B ). Kleinerman and colleagues[12] updated this series to focus on secondary sarcomas and found that there was a significantly increased risk of sarcomas within the RT field as well as an increased risk outside of the RT field, with a 13% cumulative risk of sarcoma 50 years after RT. Leiomyosarcomas were frequently diagnosed over 30 years later, indicating that older carriers remain at risk. Fletcher and colleagues[13] described the long-term risks of SMNs in retinoblastoma survivors born before 1950, a group that was not routinely treated with high-dose RT. The cumulative incidence of SMNs from age 25 to 84 was 69% in the hereditary cases and 48% in the sporadic cases, but the majority were epithelial cancers and not sarcomas. These studies indicate that retinoblastoma carriers have both an inherent risk of other SMNs and a RT-dependent risk of secondary sarcomas, which affects how these patients should be followed into adulthood.

 
 

Figure 66-1  A, The cumulative incidence of second malignant tumors is increased in children with familial retinoblastoma. B, The cumulative risk of second malignant tumors is increased in children with familial retinoblastoma who are treated with radiation therapy.  (Data from Wong GL, Boice JD Jr, Abramson DH, et al: Cancer incidence after retinoblastoma: radiation dose and sarcoma risk. JAMA 1997;1262:278.)

 

 

 

Li-Fraumeni syndrome consists of sarcoma diagnosed in the proband before age 45 years, with additional cancers (frequently soft-tissue sarcoma or breast cancer) diagnosed in other children and young adults within the family.[14] The genetic defect in some families with the Li-Fraumeni syndrome was demonstrated to be a mutation within the p53 gene. [15] [16] [17] [18] Because the pattern of first and second malignant tumors in some patients with SMNs resembles the distribution that is observed within some families with Li-Fraumeni syndrome, a series of patients with SMNs was evaluated for the occurrence of mutations at this locus. Mutations were identified in 5.1% of 59 patients who were examined. [19] [20] Pediatric patients with cancer who have neurofibromatosis have a relative risk of 8.1 of developing a SMN compared with pediatric patients with cancer who do not have neurofibromatosis.[21] Future research could demonstrate that neurofibromatosis type 1 patients who develop SMNs have coexistent germline p53 mutations.

Treatment Factors

Surgery

Surgical procedures can increase the risk of subsequent malignancy. Adenocarcinoma of the colon has been reported in several patients after ureterosigmoidostomy. The incidence of adenocarcinoma in these patients was approximately 9.9 per 1000, compared with an incidence rate of 9.9 per 100,000 in the general population.[22] The majority of reported patients have undergone this procedure for treatment of exstrophy of the bladder. The median interval between ureterosigmoidostomy and the diagnosis of colon carcinoma was 22 years.[23]

Radiation Therapy

The risk of various SMNs has been linked to the use of both RT and chemotherapy. [8] [24] [25] As treatment of childhood malignancies has intensified over the past several decades, the cure rate has increased, but so has the risk of SMNs.[8] In the cohort of 13,136 patients from the Childhood Cancer Survivor Study, 59% of the nonbreast, nonskin, nonthyroid SMNs occurred within RT fields.[26] A European case-control study of 4581 survivors by Guerin and colleagues demonstrated a radiation dose-response for the excess risk of SMNs that best fit a linear model of 0.13 per Gy.[24] This study also noted that the relative risk of SMN was increased when chemotherapy and RT were delivered concomitantly.

Patients who receive neck irradiation for malignant diseases are at risk for the subsequent occurrence of thyroid malignancies. These have been reported after treatment of patients with medulloblastoma, acute lymphoblastic leukemia, and Hodgkin's disease (HD). [27] [28] [29] The incidence of thyroid cancer in survivors of HD was 0.8% (1 in 119) among children who were treated at Stanford University.[30]Malignant thyroid tumors occurred at lower RT doses than did benign lesions.[31] Ron and colleagues reported that a linear-exponential model fit the dose-response data for thyroid cancer after treatment for childhood cancer better than a linear model did.[32]

This finding is consistent with the data of Upton, who demonstrated that the dose-response relationship for some radiation-induced experimental tumors was quadratic rather than linear.[33] These data led Gray to hypothesize that the shape of the dose-response curve was the sum of two radiation-induced processes: mutation induction and cell death.[34] Radiation-associated SMNs in other sites, however, have been fit to linear models (see later discussion).

Central nervous system tumors, including meningiomas and gliomas, have been reported with increasing frequency after direct or incidental irradiation of the brain, which is not surprising when the occurrence of brain tumors in children treated with low doses of RT for tinea capitis is recalled. Neglia and colleagues[35] reported that the excess radiation-associated risk for secondary gliomas was 6.8 and that for meningiomas was 9.9. There was a linear dose response, with a steeper slope for meningiomas (1.1) than for gliomas (0.3; Fig. 66-2 ). The excess risk for gliomas was greatest in children who were irradiated before age 5. Other researchers have confirmed the increased risk of central nervous system tumors in children whose treatment for acute lymphoblastic leukemia included cranial irradiation.[36] In children with neurofibromatosis type 1, there was a threefold relative risk of second nervous system tumors after RT for optic pathway gliomas compared to unirradiated patients.[37]

 
 

Figure 66-2  The risk of subsequent glioma (closed squarespurple line) and meningioma (open squaresblue line) increases with radiation dose.  (Data from Neglia JP, Robison LL, Stovall M, et al: New primary neoplasms of the central nervous system in survivors of childhood cancer: a report from the Childhood Cancer Survivor Study. J Natl Cancer Inst 2006;98:1528.)

 

 

 

Genetic loci associated with the occurrence of Wilms’ tumor have been identified on chromosomes 11, 17, and 19. [38] [39] [40] Some patients have germline mutations in WT1, the only Wilms’ tumor-associated gene that has been sequenced. [41] [42] Li and associates reported that the frequency of SMNs in a cohort of successfully treated Wilms’ tumor patients was 6% 20 years after diagnosis.[43] SMNs were diagnosed only in irradiated patients. Breslow and colleagues reviewed the occurrence of SMNs among patients entered on the National Wilms Tumor Studies.[44] The cumulative risk of SMN was 1.6% 15 years after diagnosis ( Fig. 66-3 ). The relative risk of developing SMN was increased in patients who had received RT, the relative risk increasing with increasing radiation dose. Administration of doxorubicin increased the relative risk at each level of radiation exposure.[44] Other researchers have reported a cumulative frequency of SMN after treatment for Wilms’ tumor at 3.9% at 20 years, with a relative risk of 11.0.[45] Patients with bilateral Wilms’ tumor were not at increased risk for SMNs, according to the National Wilms Tumor Study Group analysis.[44]

 
 

Figure 66-3  Children who have been successfully treated for Wilms’ tumor have a significant risk of developing a second malignant tumor.  (Data from Breslow NE, Takashima JR, Whitton JA, et al: Second malignant neoplasms following treatment for Wilms’ tumor: a report from the National Wilms Tumor Study Group. J Clin Oncol 1995;13:1851.)

 

 

 

Sarcomas of bone have been reported both in patients with hereditary retinoblastoma and in those who survive other types of childhood cancer. The cumulative risk of SMN in bone was estimated to be 2.8% among 9170 patients who were evaluated but was 14.1% among those with retinoblastoma and 22.1% among those who had been treated for Ewing sarcoma (ES) at 20 years after diagnosis ( Fig. 66-4).[46] The relative risk was 2.7 among patients whose treatment had included RT; the relative risk increased with increasing RT dose and more intensive use of alkylating agents.[46] Other researchers reported a relative risk for osteosarcoma of 88–1515 in patients who had been treated for retinoblastoma; the relative risk rises to 800 after treatment for ES. [3] [47] Hawkins and colleagues calculated the cumulative frequency of bone cancer in previously irradiated childhood cancer survivors as 0.5% among those who were not treated for retinoblastoma and as 7.2% among those who were treated for heritable retinoblastoma.[48] A dramatically high risk of sarcoma of bone after treatment of ES was observed in a multiinstitutional study.[49] The cumulative frequency of a SMN in ES patients who were treated successfully was 9.2% at 20 years after diagnosis, and that of a secondary sarcoma was 6.5%. No secondary sarcomas developed in patients who had received less than 48 Gy.[49] A case-control study of bone sarcoma as a SMN found no difference in relative risk between patients treated with orthovoltage and megavoltage RT.[46] A decreased risk might have been expected because of the lower absorbed doses in bone after high-energy RT. Successfully treated patients are at risk of developing carcinomas (e.g., of the skin) within prior RT treatment volumes given at a very early age.[50]

 
 

Figure 66-4  The cumulative incidence of bone sarcoma as a second malignant tumor is increased in children who were treated for retinoblastoma compared with those who were treated for other forms of childhood cancer.  (Data from Tucker MA, D'Angio GJ, Boice JD Jr, et al: Bone sarcomas linked to radiography and chemotherapy in children. N Engl J Med 1987:317:588.)

 

 

 

Irradiation of breast tissue increases the risk of breast cancer, as was demonstrated in women who were exposed to diagnostic radiograph for pulmonary tuberculosis and survivors of the atomic bomb detonations in Hiroshima and Nagasaki. [51] [52] The reported relative risks of breast cancer in girls who were treated for HD have varied widely. The largest study with the longest follow-up reported a relative risk of 11.5.[53] In this large, population-based study of 383 HD survivors, all 16 women who developed breast cancer had supradiaphragmatic irradiation. The cumulative risk of breast cancer at 25 years was 9.9% for all HD patients and 12.2% for those who were treated with supradiaphragmatic RT.

Total body irradiation, a component of most preparative regimens for allogeneic bone marrow transplantation for malignant diseases, is associated with a cumulative risk for the occurrence of a second solid neoplasm of 6.1% at 10 years after treatment.[54]

Hormone Therapy

Growth hormone (GH) treatment is necessary for the management of some children who received cranial irradiation as part of their therapy for acute lymphoblastic leukemia or brain tumors. The question of whether use of GH increases the risk of second cancers was addressed in the Childhood Cancer Survivor Study cohort.[55] Using a time-dependent Cox multivariate model, the authors found that GH increased the risk of SMNs by 2.1, which was less than previous estimates. The majority of SMNs were intracranial, especially meningiomas. Although the use of radiation was taken into account in their model, the radiation dose was not, so it is possible that the need for GH was a surrogate for radiation dose. The possible risk identified by these two studies must be weighed in the context of the substantial benefits that accrue to these patients as the result of GH therapy, including improved linear growth and bone mineral accretion.

Chemotherapy

The significance of prior treatment with chemotherapy in the pathogenesis of SMNs was first evaluated in detail in cohorts of adults who had been treated successfully for HD. The risk factors for the occurrence of SMNs in pediatric patients after treatment for HD have been evaluated less thoroughly.

Bhatia and coworkers reported that the cumulative risk of developing any SMN after treatment for HD in childhood was 7% at 15 years after diagnosis.[56] The risk of developing non-Hodgkin's lymphoma (NHL) was 1.1%, and the risk of developing any type of leukemia was 2.8% at 15 years after diagnosis. The investigators found that the relative risk of leukemia was proportional to the alkylating agent dose score, that is, the cumulative dose of those drugs. The actuarial risk of developing acute myelogenous leukemia 10 years after diagnosis was 11% among pediatric patients treated at Stanford University with low-dose (2500 cGy) RT and MOPP (nitrogen mustard [M], vincristine [O], procarbazine [P], and prednisone [P]) chemotherapy, which has an alkylating agent dose score of 2.[57] The risk was 1.1% 15 years after diagnosis among pediatric patients treated with involved or extended field RT and various chemotherapy regimens that did not contain nitrogen mustard (vincristine, prednisone, doxorubicin, with or without procarbazine; cyclophosphamide, vincristine, prednisone, procarbazine, or methotrexate).[58]

Other investigators have studied the risk of developing SMNs after various chemotherapeutic agent exposures. Tucker and colleagues reported that prior treatment with an alkylating agent increased the risk of developing bone cancer or leukemia ( Fig. 66-5 ) as SMN.[46] De Vathaire and coworkers demonstrated that dactinomycin increased the risk of a bone or soft tissue SMN (relative risk: 8.7).[59] Garwicz and associates reported that treatment with classical alkylating agents (nitrogen mustard, cyclophosphamide, lomustine), nonclassical alkylating agents (procarbazine), vinca alkaloids (vinblastine, vincristine), or prednisone each increased the relative risk of SMN.[21] Only procarbazine increased the relative risk of SMN when it was included in a two-factor multivariate model.[21] Klein and colleagues reported an increased relative risk for SMN with higher doses of several agents, including cyclophosphamide (RR 6.3 for doses >8000 mg/m2), cisplatinum (relative risk: 2.8 for doses >435 mg/m2), and 6-mercaptopurine (relative risk: 4.5 for doses >5000 mg/m2).[60] Neglia and associates reported that treatment with an anthracycline (relative risk: 1.51 for doses of 101–300 mg/m2; relative risk: 1.44 for doses >300 mg/m2) or epipodophyllotoxin (relative risk: 2.78 for doses >4001 mg/m2) did not demonstrate an increased relative risk with increasing alkylating agent dose score or cisplatinum exposure.[6]

 
 

Figure 66-5  The cumulative incidence of leukemia as a second malignant tumor is increased in children who were treated for Hodgkin's disease compared with those who were treated for other forms of childhood cancer.  (Data from Tucker MA, Meadows AT, Boice JD Jr, et al: Leukemia after therapy with alkylating agents for childhood cancer. J Natl Cancer Inst 1987;78:459.)

 

 

 

The epipodophyllotoxins have been identified as important leukemogens. Pui and coworkers reported the risk of secondary acute myelogenous leukemia (AML) as 4.7% at 6 years after diagnosis among patients treated for acute lymphoblastic leukemia.[61] The risk was substantially higher (19.1%) among patients with T-cell leukemia. These investigators subsequently demonstrated that the risk of secondary AML in this population was related to the administration of epipodophyllotoxins, with the cumulative frequency of AML being 12.3% among those who were treated twice weekly and 12.4% among those who were treated weekly, compared with a frequency of 1.6% among those who were treated with the drug less frequently or not at all ( Fig. 66-6 ).[62]

 
 

Figure 66-6  The cumulative incidence of acute myelogenous leukemia after treatment for acute lymphoblastic leukemia is increased among children whose therapy includes weekly epipodophyllotoxin.  (Data from Pui C-H, Behm FG, Raimondi SC, et al: Secondary acute myeloid leukemia. N Engl J Med 1989;321:136.)

 

 

 

The risk of secondary leukemias after the treatment of solid pediatric tumors has been linked to treatment with radiation, epipodophyllotoxins, and vinca alkaloids.[25] The onset of secondary leukemia isgenerally considered an early event, but Haddy and coworkers noted peaks in the first 10 years and again after 20 years.[25] The risk of secondary AML depends on the cumulative dose of drug administered and the schedule of administration; the frequency was reported to be 0% among germ cell tumor patients who were treated with less than 2000 mg/m2, 5.9% among childhood acute lymphoblastic leukemia patients who were treated with 1800 to 9900 mg/m2, 11.3% among germ cell tumor patients who received more than 2000 mg/m2, and 18.4% among pediatric NHL patients who received 4200 to 5600 mg/m2. [63] [64] [65] Hawkins and associates reported an increased risk of secondary leukemia with increasing cumulative dose of epipodophyllotoxin.[66]

The carcinogenic potential of the anthracycline doxorubicin was suggested by the results of a previous case-control study of risk factors for leukemia as a SMN. Increasing doxorubicin dose was found to be associated with an increasing relative risk of leukemia as a SMN after adjustment for the cumulative dose of alkylators given concomitantly.[67] More recently, prior treatment with anthracyclines or epipodophyllotoxins has been shown to increase the risk of therapy-related acute promyelocytic leukemia and of secondary leukemia or myelodysplasia. [68] [69] Patients who received 1.2 to 6.0 g/m2 of epipodophyllotoxin had a relative risk of developing leukemia of 3.9, whereas those who received more than 170 mg/m2 of anthracycline had a relative risk of developing leukemia of 3.0.[68] This finding is of interest, as doxorubicin is now known to have topoisomerase II as one of its targets—the same target as that of the epipodophyllotoxins. A case-control study of risk factors for bone sarcoma as a SMN did not identify any effect of doxorubicin therapy on the risk of developing such SMNs.[46] An analysis of risk factors for any SMN in a large cohort of pediatric patients with cancer demonstrated that treatment with doxorubicin was the only factor that was identified (other than treatment with carmustine) that increased the risk of a SMN. (This finding apparently extended the earlier suggestion that doxorubicin was leukemogenic.[2]) The evidence in adults will be reviewed later in the chapter.

The leukemogenicity of topoisomerase II inhibitors could be related to the high degree of specificity of these agents for specific DNA targets, including the myeloid-lymphoid leukemia gene and the acute myeloid leukemia 1 (AML1) gene. [70] [71] [72]

Although most studies of carcinogenicity of chemotherapeutic agents have focused on the development of leukemia after treatment, it is clear that solid tumor induction is also possible after exposure to one or more chemotherapeutic agents. The best example of this is the occurrence of solid SMNs in genetically predisposed retinoblastoma patients who were treated only with cyclophosphamide after enucleation.[73] As our ability to identify genetically predisposed patients improves, our understanding of the apparent anomaly of solid tumor induction after systemic exposure to a carcinogenic agent will increase.

Immune Suppression

Immune suppression is a component of allogeneic bone marrow transplantation. To prevent graft-versus-host disease, antithymocyte globulin can be administered to the recipient, or the bone marrow can be manipulated to remove T cells. These manipulations and bone marrow transplantation from an unrelated bone marrow donor increase the risk of Epstein-Barr virus-associated B-cell lymphoproliferative disorder. The cumulative incidence rates at 10 years after bone marrow transplantation were 11.3% among those who were treated with antithymocyte globulin, 11.4% among those who received T-cell-depleted bone marrow, and 2.3% among those who received bone marrow from unrelated donors.[74]

Recommendations

Survivors of childhood cancer are at increased risk of developing a new malignancy. The magnitude of this risk is modulated by several factors, including the patient's genetic susceptibility, the type of surgical procedure that is employed for removal of the tumor, the use of RT as part of the treatment plan, the chemotherapeutic agents that are employed, the severity of immune suppression present at the completion of all treatment, and environmental exposures.[75]

All former patients should remain under a physician's care indefinitely and should undergo frequent physical examinations, preferably by a physician or other health care worker who is familiar with the problem of therapy-induced malignancy.

Patients who have been irradiated should have plain radiographs obtained whenever local pain occurs in a previously irradiated bone. Patients who have received irradiation to volumes that include the breast, uterine cervix, or intestine should undergo routine evaluation with available screening tests, such as mammography, Pap smear, and stool examination for the presence of occult blood. Annual mammography should be initiated no later than 10 years after breast irradiation. Careful physical examination of irradiated patients will facilitate the early identification of thyroid nodules and skin cancer.

All patients who have been treated with an alkylating agent, procarbazine, or a topoisomerase II inhibitor should have a complete blood count every 6 to 12 months for a minimum of 12 years after diagnosis. The presence of macrocytosis and/or cytopenia should prompt evaluation of the bone marrow.

The adverse health consequences of tobacco use, including carcinogenesis, are well documented. Unfortunately, childhood cancer survivors are only slightly less likely to use tobacco. [76] [77] Physicians should counsel childhood cancer survivors about the potential for the adverse effects of tobacco use to interact with the adverse effects of their prior therapy, such as irradiation of the oropharynx, esophagus, and/or lung.

ADULTS

Genetic Factors

Travis provides an excellent, comprehensive overview and update of research concerning SMNs focusing on solid tumors in the adult age group.[78] She stresses that several factors besides chemotherapy, RT, and their interplays need to be considered. For example, the complex potential interactions of alkylating agents, RT, and tobacco smoking are developed extensively in her discussion of lung cancers as SMNs, especially in HD patients. She points out the additional variable in HD patients (i.e., impaired immunologic responses), so that data derived from long-term survivors of HD cannot be extrapolated with confidence to patients with other SMNs. Studies of second cancers can help point to common genetic mechanisms. For example, a study of secondary pancreatic cancers showed an increased propensity of pancreas cancer to occur not only after tobacco-associated cancers, but also after breast cancer (male and female) and ovarian cancer, which are associated with BRCA2 mutations.[79]

The potential importance of immunodeficiency in the appearance of SMNs was the hypothesis explored by Hemminki and coworkers in an imaginative epidemiologic investigation.[80] They noted that skin cancers and NHLs were the most frequent malignant lesions to appear in immune-deficient patients (e.g., in renal transplant patients who were given needed post-transplant immune-suppressing agents). Using the Swedish nationwide database, they found 4301 secondary skin cancers and 1672 NHLs in a period of about 40 years among 10.2 million survivors of a primary cancer. The researchers found increased risks—up to 12-fold depending on patient sex and primary tumor site—for these two neoplastic entities. These results suggest that immunodeficiency could play a role in the appearance of SMNs.Several sites of primary tumors in both males and females were encompassed in their study, making therapy-induced immunosuppression unlikely. This conclusion points to possible underlying genetic factors as contributory influences ( Box 66-1 ).

Box 66-1 

EVALUATION FOR PATIENTS AT RISK FOR SECOND MALIGNANT NEOPLASMS

Risk Factors

  

   

Treatment with radiation therapy, an alkylating agent, and/or a topoisomerase II inhibitor

  

   

History of hereditary retinoblastoma (unilateral with a positive family history or bilateral)

  

   

Carrier of ataxia-telangiectasia

  

   

Postsurgical chronic lymphedema

  

   

Ureterosigmoidostomy

  

   

Estrogen treatment

History

  

   

Pain in any previously irradiated area

  

   

Bruising

  

   

Gum bleeding

  

   

Pallor

  

   

Easy fatigability

  

   

Breast lump

  

   

Cough

  

   

Chest pain

  

   

Hemoptysis

  

   

Blood in stool

  

   

Constipation

  

   

Tenesmus

  

   

Hematuria

  

   

Increased urinary frequency/incontinence

  

   

Difficulty voiding

  

   

Intermenstrual bleeding

Physical Examination

  

   

Pallor

  

   

Petechiae

  

   

Chronic skin ulceration

  

   

Presence of lump (mobility, tenderness, consistency)

  

   

Asymmetric breath sounds

  

   

Nodule in prostate

  

   

Abnormal uterine cervix

Evaluation

  

   

Plain radiographs of any painful area or mass in a previously irradiated area

  

   

Stool examination for occult blood (any patient who received any abdominal irradiation)

  

   

Pap smear

  

   

Urine cytology (in any patient with hematuria and a history of bladder irradiation and/or treatment with a cyclophosphamide or ifosfamide)

  

   

Mammogram

  

   

Additional tests as indicated by the history and physical examination

The importance of the Li-Fraumeni syndrome as a risk factor for cancer in young adults has been discussed previously.[75] Cancer families provide evidence supporting laboratory studies of the genetic bases of adult cancers. Cancer predisposition genes for breast cancer (BRCA1, BRCA2), colon cancer (MSH1, MSH2, APC, DCC), malignant melanoma (CDK2), and renal cell carcinoma (RCC) have all been identified.

At the clinical level, carriers of the ataxia-telangiectasia gene (ATM) are more likely to develop cancer; Swift and colleagues estimate a 3 to 4 excess relative risk of cancer in male and female carriers.[81]Female breast cancer was the most frequent cancer reported (excess risk of about 5) and was more likely to occur in those exposed to ionizing radiation. Little and coworkers compared the relative risks of developing post-therapeutic irradiation SMNs to those among Japanese survivors of the atomic bomb blasts.[82] They found the relative risks of developing leukemia or lung, bone, or ovarian cancer to be higher among the Japanese survivors than among treated patients. Neither chemotherapy nor underlying genetic factors seemed to play a role in their results.

The cytogenetics of SMNs have been the subject of intensive investigation. Le Beau and associates described characteristic abnormalities of chromosome 5(del (5q)) and 7(del (7q)) in patients with treatment-associated acute myeloid leukemia.[83] (The terms acute myeloid leukemia [AML] and acute nonlymphoblastic leukemia will be used interchangeably in this discussion. In most cases, the term that is used will be the one employed by the authorities being cited.) Detourmignies and colleagues found t(15;17) in treatment-associated acute promyelocytic leukemia (t-APL) and the other forms of acute myeloid leukemia (t-AML), the same translocation as is found in those diseases de novo.[84] In addition, t(8;21), t(9;11) and inv[16] were found in t-APL and other t-AMLs. They tend to arise in patients with solid tumors, have short latent periods, and are associated with prior therapy with drugs that inhibit topoisomerase II. Cytogenetic evaluations of solid SMNs have demonstrated chromosome 22 deletions in meningiomas and deletions in chromosomes 10 and 17 in malignant astrocytomas. [85] [86] These cytogenetic findings could have important implications regarding early identification and early treatment of SMNs.

Treatment Factors

Surgery

In adults, perhaps the most common SMN ascribed to surgery is angiosarcoma in a lymphedematous structure (Stewart-Treves syndrome), a complication that has been described most often after radical mastectomy.[87] The tumors tend to arise in the edematous arm rather than in irradiated areas.[88] Marchal and coworkers conducted a survey of breast angiosarcomas that develop in women who were treated with breast-conserving techniques.[89] They found 9 cases among almost 20,000 women but found no conclusive evidence linking therapy as a causative factor.

Radiation Therapy

Lindsay and colleagues have reviewed the factors involved in radiation carcinogenesis.[90] Breast, thyroid, bone, soft tissues, and organs and tissues that are prone to develop SMNs after RT given in childhood are also vulnerable to radiation oncogenesis when treatment is given during the adult years ( Fig. 66-7 ). The dose-response relationship has classically been described as bell-shaped: an initial increase with low and moderate radiation doses to a peak followed by a rapid decrease at higher doses due to killing of the vulnerable cells. The observed rate for second cancers after radiation of Hodgkin's disease has been higher than predicted, and Sachs and Brenner proposed a model using normal cell repopulation to account for high cancer rates after higher radiation doses.[91] Precise RT dose-response relationships remain murky, in part owing to limited data about the true RT doses at observed SMN sites.

 
 

Figure 66-7  Schematic illustration of risk factors for second primary cancers. ANLL, acute nonlymphoblastic leukemia; CLL, chronic lymphoblastic leukemia.  (Data from Storm HH: Second primary cancer after treatment for cervical cancer. Later effects after radiotherapy. Cancer 1998;61:679.)

 

 

 

LYMPHOMA.

Survivors of Hodgkin's disease are at increased risk for a variety of cancers, especially lung, colorectal, and breast cancer, all of which were more common in those who were treated with chemotherapy and radiation.[92] In fact, a leading cause of death among long-term survivors of HD is second cancer. Treatment of HD has served as a model for therapeutic radiation-induced carcinogenesis, especially the risk for breast cancer. [91] [93] [94] Wendland and colleagues studied secondary breast cancers in 8036 females with HD from Surveillance, Epidemiology and End Results (SEER) registries. Second breast cancers were seen in 2.3%, and the standardized incidence ratio was 1.9 for women treated with RT.[94] Interestingly, the breast cancer-free survival curves crossed for irradiated and nonirradiated patients. There was a paucity of early breast cancers in the radiated group, but an increased incidence over time. This was modeled best with a nested proportional hazard model, possibly owing to a therapeutic effect on pre-existing tumors combined with induction of latent tumors. Travis and colleagues[93] studied 3817 female 1-year survivors of HD and noted an age- and dose-dependent cumulative risk of breast cancer development that was lower in patients who were treated with alkylating agents. In their model, women who were treated at age 25 would have an estimated cumulative absolute risk of breast cancer of 1.4% by age 35, 11% by age 45, and 29% by age 55. Projections from these studies should be taken with caution, however, owing to decreases in the dose and volume radiated in the modern era.

A SEER study of 77,876 patients with NHL showed that irradiated patients had a similar risk of SMNs compared with unirradiated patients (relative risk: 1.04).[95] The SMN types differed, with more sarcomas, breast cancers, and mesotheliomas occurring in irradiated patients. Younger patients had an increased relative risk for SMNs, which was more pronounced in irradiated patients.

BONE AND SOFT TISSUE.

The largest follow-up study of second cancers in patients with soft-tissue sarcoma primaries (N = 6,671) comes from the Swedish Family-Cancer Database, which identified 650 second cancers (9.7%).[96]The median time to second cancer was 7 years, and the standardized incidence ratio was 1.42 (95% CI: 1.31 to 1.53). The most common second cancer was another soft-tissue sarcoma. Although 10 of 39 were at the same primary site, the majority occurred at different anatomic sites. A weakness of this database is the lack of treatment information, so Ji and colleagues did a reverse analysis, which showed an increased rate of second soft-tissue sarcomas after any primary tumor as well (1.92, 95% CI: 1.78 to 2.07).[96] This suggests that factors other than treatment, such as genetic predisposition and environmental risk, also play an important role.

Lagrange and coworkers reported 80 radiation-associated sarcoma cases collected by a consortium of French cancer centers.[97] The median dose of RT was 50 Gy (range: 9 to 110 Gy) delivered to adults (median age: 44 years) for a variety of primary diagnoses. Of the histologically proven secondary sarcomas, 70% were bone and 30% were soft-tissue sarcomas, osteosarcomas and malignant fibrous histiocytomas predominating among them. Unlike the earlier study by Tountas and colleagues,[98] these authors could not demonstrate a correlation with increasing dose. The outcomes were poor despite aggressive treatment based largely on surgical maneuvers.

HEAD AND NECK.

In the head and neck, second primary tumors are commonly observed as a result of premalignant field changes due to environmental risk factors, such as tobacco. A multicentric, case-control study of laryngeal and hypopharyngeal cancers showed an average second cancer rate of 2.1% per year. A higher risk was associated with tobacco, alcohol, and butter consumption, whereas citrus fruit consumption was protective.[99] An attempt has been made to prevent second cancers with antioxidant therapy in a randomized, placebo-controlled trial.[100] Unfortunately, α-tocopherol did not lower the overall recurrence/second cancer rate and actually increased the rate in the first 3.5 years.

In an analysis of 326 consecutive patients with nasopharyngeal carcinoma who were treated with definitive RT, 5.2% developed second primary cancers at a rate of 1% per year.[101] Only about one third of these were in-field, and there was not an association with the total prescribed dose. The only second tumors that occurred after 5 years were within the radiation field, supporting the concept of a lag time for radiation-induced solid cancers.

GYNECOLOGIC.

Storm reviewed the frequency of SMNs in a cohort of 24,970 Danish women with invasive cervical cancer and 19,470 who had carcinoma in situ followed for 30 or more years.[102] Taken together, there was an increased relative risk of 1.9 in irradiated patients who survived 30 or more years, representing an excess of 64 cases per 10,000 women annually (see Fig. 66-7 ). Werner-Wasik and coworkers analyzed the frequency and patterns of SMNs in women with cervical cancer and came to a different conclusion.[103] They found 11 SMNs among 10 of the 125 women with FIGO stage I and II cervical carcinoma who received RT in a recent 10-year period (1980 to 1990). All of the SMNs were outside the fields of irradiation, and none of the women had received chemotherapy. The researchers concluded that the increased relative risk of a SMN might be genetically based, as the administered treatments did not appear to be factors. The Werner-Wasik report again emphasizes the need to consider the multiple possible contributing factors to oncogenesis (see Fig. 66-7 ).

Sturgeon and associates found an excess of SMNs among women with vulvar or vaginal cancers.[104] Most of the SMNs were smoking-related (lung, upper airways) or in patients infected with the human papillomavirus, which is known to be associated with cancers of the genital tract. A related observation by Hemminki and Dong is of interest; they found an increase in anal cancers in both women with cervical cancer and their husbands, implicating human papillomavirus in the etiology of anal cancer.[105]

Hall and colleagues found an increased risk of ovarian cancer as a SMN in certain sets of women.[106] These were women younger than 50 years of age with melanoma or cancer of the colon, breast, cervix, uterine corpus, or ovary. The relative risks ranged from about 5 to almost 20.

Bergfeldt and coworkers could not substantiate a reputed increase in breast cancers among women with ovarian cancer.[107] Their case-control study from a pool of 5060 Swedish women led them to conclude that increased surveillance (mammography) of surviving patients with ovarian cancer was not warranted. Travis and associates likewise attributed secondary breast cancers to factors other than therapy, although they attribute soft-tissue, bladder, and rectal malignancies to RT and leukemia to chemotherapy.[108]

Buiatti and colleagues reported a population-based study of second primary cancers.[109] They considered only the 463 metachronous SMNs that developed among the 19,252 adults with primary cancers of the colon, rectum, lung, stomach, and female breast who constituted the study population. Significantly higher risks of developing another cancer were found in patients under 65 years of age. Associations of three types were found:

  

1.   

Between primary rectal and secondary kidney cancers

  

2.   

Between colon and later ovarian malignancies

  

3.   

Between female breast and subsequent rectal cancers, although cancer in the opposite breast constituted the highest risk in this group

No correlations were made between the treatments that were used for the primary tumor and the secondary cancer. This report, together with that of Werner-Wasik and associates,[103] highlights the fact that SMNs should not all be assumed to have an iatrogenic basis.

GENITOURINARY TRACT.

The incidence of second neoplasms resulting from RT for prostate cancer is controversial. Movsas and associates found no increase in the risk of SMN among 543 of their patients when compared with a matched set of 18,135 men derived from the Connecticut Tumor Registry.[110] Most of the SMNs developed outside the RT fields and were associated with lifestyles that were predisposing to cancer. Johnstone and colleagues also found no definite increase.[111] Groups using the SEER database have found conflicting results. [112] [113] [114] [115] In comparison to patients who were treated only with surgery, there was a small but significantly increased risk of in-field tumors, especially bladder and rectal cancers, following RT. [112] [113] However Moon and colleagues also found an increased risk of second cancers outside the field and in men who were treated with transurethral resection of prostate alone.[115] These results suggest the researchers did not adjust for all relevant factors. In fact, when Cox modeling was used and adjusted for attained age, no significant excess risk from RT was found. [114] [116]

SMNs occurring in men with testicular cancers have been studied by several investigators. Wanderas and associates found an increased risk of second germ cell cancer, usually of the same histology, among 2201 patients, the risk being highest among men younger than 30 years of age at first diagnosis.[117] More SMNs than expected were found by Ruther and colleagues in their multicenter collection of men with pure seminomas.[118] These included both nontesticular and testicular SMNs. Other researchers, who have studied larger numbers of patients, have documented increased relative risks for gonadal and nongonadal carcinomas of 1.2 to 2.3. Higher risks were recorded for leukemia (relative risk: 2.4) and soft-tissue sarcomas (relative risk: 3.0).[119] The SMNs tended to develop in irradiated fields.

Travis and coworkers showed that men who were treated for testicular cancer had a persistent increased risk of second cancers (relative risk: 1.9) in a large cohort study of 40,576 survivors.[120] SMNs of the lung, colon, bladder, pancreas, and stomach accounted for almost 60% of the excess cancers. The risks of cancer 40 years after treatment were 36% for seminomas and 31% for nonseminomas compared to 23% for the general population to age 75. Richiardi and associates also found an elevated risk of second cancers with a standardized incidence ratio of 1.65 compared to population controls in a cohort of 29,511 patients with testicular cancer.[121] Interestingly, they found a markedly increased risk of myeloid leukemia, especially in nonseminomas that were diagnosed after 1990 (standardized incidence ratio: 38). The elevated risk in these patients could be due not only to genetic and environmental factors, but also to outdated RT techniques in which extended fields were used even for early-stage disease.[122]

BREAST CANCER.

After a diagnosis of breast cancer, the risk of second cancers is increased in general in comparison to population controls. A study of 335,191 women with either invasive or noninvasive breast cancer revealed a second primary rate of 12% for women who were diagnosed before age 50 and 17% for women who were diagnosed over age 50.[123] Noteworthy is the fact that the standardized incidence ratios for second cancers actually decreased with age, suggesting that genetic mechanisms play a role. The patterns of second cancers, namely, second breast, bone, colorectal, sarcoma, leukemia, lung, ovarian and thyroid cancer, have a shared pattern of risks with BRCA1, BRCA2, p53, and PTEN mutations.[123] In a study of 491 female carriers of either BRCA1 or BRCA2 mutations, the risk of contralateral breast cancer was 29.5% at 10 years, with BRCA1 carrying a higher hazard than BRCA2.[124] Studies of second malignancies after male breast cancer have similarly shown increased rates compared to population controls, especially breast, gastrointestinal, and prostate cancers, but unlike female breast cancer, the risks increased over time. [125] [126]

In women who had been treated uniformly with wide local excision, axillary dissection, and postoperative irradiation, the cumulative 10-year SMN rate was 16% in the 1253 women who were reviewed, about half of whom had second breast malignancies.[127] A 20-year analysis of 1801 similarly treated women showed a contralateral breast cancer rate of 15.4%, the majority (83%) of which were the invasive type.[128]

The contralateral breast does receive some radiation dose during breast RT. It is in the range that is known to be carcinogenic, especially in young women.[129] A case-control study of 41,109 women with breast cancer showed an attributable contralateral breast cancer risk of only 2.7% from prior irradiation.[130] Some of the more important variables that were considered in the analyses are shown in Table 66-1 . An increase in relative risk was observed in irradiated women surviving 10 or more years, but only in those who had been treated when under 45 years of age (relative risk: 1.85). The risk among this sample increased with increasing RT dose, the analyzed range extending from 1.99 Gy to 4 Gy, and the relative risks from 1.54 to 2.35, respectively (P = 0.003).


Table 66-1   -- Relative Risk of Contralateral Breast Cancer after Radiation Therapy

Age at Treatment (10-Year Survivors)

No. Exposed

Total No.

Relative Risk (95% Confidence Interval)

<45 yrs old

45

143

1.85 (1.15–2.97)

>45 yrs old

55

298

1.08 (0.74–1.57)

Data from Boice JD, Harvey EB, Blettren M, et al: Cancer in the contralateral breast after radiotherapy for breast cancer. N Engl J Med 1992;326:781.

 

 

 

The use of adjuvant RT appears to increase the risk of both contralateral breast and other cancers (skin, lung, colon, and esophagus) in some, [131] [132] [133] [134] but not all studies.[135] Additional evidence for an increased risk of esophageal cancer after breast cancer has been found in population-based studies. [136] [137] This may be attributable to the radiation field that was used to include the internal mammary nodes. Although the absolute risk from RT of breast cancer is small, the doses to the contralateral breast and mediastinum can and should be reduced by using modern RT techniques.

NERVOUS SYSTEM.

Jones conducted an extensive retrospective analysis of brain tumors occurring in patients after RT for pituitary lesions, a unique group that receives relatively high doses of RT for nonmalignant lesions.[138]He concluded that the risk of RT oncogenesis in adults is low after small-field RT in the doses usually used for the control of pituitary disorders. When both irradiated and unirradiated adults with pituitary tumors who developed regional fibrosarcomas, gliomas, and meningiomas were compared, the role of therapeutic RT in the genesis of such SMNs became questionable. Erfurth and associates not only find no clear evidence implicating RT, but also again raise the possibility that underlying genetic factors are responsible.[139]

LEUKEMIA.

The risk of secondary leukemia is greatest within the first several years after radiation, whereas the risk for solid tumors generally becomes evident thereafter. Although relatively rare, an excess absolute risk of AML was detected in patients who have been treated for HD.[140] The excess risk was greatest in the first 10 years after treatment and subsequently remained elevated. In cervical cancer patients, Storm described a significantly increased relative risk of acute nonlymphoblastic leukemia (relative risk: 3.5), manifest in the early postirradiation years but not of chronic lymphoblastic leukemia (CLL).[102] The relative risk decreased by the 10th year after diagnosis, and the incidence approximated that of the general population by the 15th year after diagnosis. This was unlike the experience with solid tumors, in which relative risks continued to increase with time.

Intensity-Modulated Radiotherapy and Proton Radiotherapy

Technological advancements in RT over the past several decades have led to safer, more tolerable treatment. Intensity-modulated radiotherapy (IMRT) has been popularized in recent years owing to the ability to shape the irradiated volume in three dimensions. This permits both a limitation of doses to adjacent normal structures and an increase in the dose to critical diseased sites that need a boost, allowing for dose escalation opportunities. A disadvantage of IMRT is higher integral body doses, resulting both from the spreading of a low dose to a greater volume and from longer treatment times with the attendant increase in incidental irradiation due to machine head leakage. Higher integral doses have led to concerns about a possible increased risk of fatal SMNs; indeed, these have been calculated to be 1.7% for conventional treatment compared to 2.1% for IMRT.[141] These worries were echoed by Hall, who also pointed out a potentially increased risk of SMNs from proton RT due to neutron contamination of the proton beam.[142] Proton RT has the advantage of no exit dose, but the most common method of proton field generation, passive scattering, is associated with a higher level of neutron contamination. Active scanning proton RT has less neutron contamination than passive scattering and is predicted to deliver a lower integral dose than photon IMRT. Hall's neutron contamination calculations and SMN risk for passive scattering proton RT were widely criticized as being based on outdated equipment.[143] The point still stands, however, that active scanning proton RT may provide the best conformality with the lowest risk of SMNs, despite a very high price tag.[144]

Radionuclide Therapy

Unlike external beam RT that is aimed to a particular site, radionuclide therapy is delivered intravenously and any localization is a result of specific affinities. Although the use of radioimmunotherapy for lymphoma with radiolabeled anti-CD20 antibodies is on the rise, the most experience with radionuclide therapy has been with I-131 for thyroid cancer. In a pooled European cohort of patient with primary thyroid cancers, patients who were treated with I-131 were found to have an increased incidence of bone, soft-tissue, colorectal, and salivary gland cancers.[145] A multinational record linkage study of 39,002 patients with primary thyroid cancer demonstrated an increased risk of second cancers both after and before treatment for the thyroid cancers.[146] A similar observation was seen in both the SEER database and a cohort from Leiden University Medical Center, indicating that common genetic and environmental risk factors are probably more important than I-131 causation. [147] [148]

Chemotherapy

LEUKEMIA–LYMPHOMA.

Most of the SMNs reported after chemotherapy have been AML or NHL. The alkylating agents were the first to be implicated in the etiology of these SMNs and remain the most frequently implicated agents. [149] [150] Subsequently, the leukemogenicity of the nitrosoureas was recognized.[151] Greene and colleagues estimated the increased relative risk of AML in patients with brain tumors who had been given carmustine to be about 25.[152] More recently, other agents have been associated with AML. These include the epipodophyllotoxins, the platinum compounds, and their combinations. [25] [153]

AML that developed in women with ovarian cancer who had been given alkylating agents was one of the first iatrogenic chemotherapy-related SMNs to be reported in convincing numbers.[150] The relative risk reported by Reimer and associates was more than 170 for women who received alkylating agents compared with those who did not. That estimate was updated by Greene and colleagues, who found a relative risk of 110 and the excess risk of acute nonlymphoblastic leukemia to be 5.8 cases per 1000 women per year.[154] Melphalan and chlorambucil were the two agents that were most strongly implicated (relative risks of 122 and 159, respectively), the risk of leukemia being dose-related.

Kaldor and coworkers published an international case-control study in 1990 of women with ovarian cancer and reported increased risks of leukemia in those who had received cyclophosphamide, chlorambucil, melphalan, thiotepa, or dihydroxybusulfan (treosulfan) as single agents.[155] The relative risks, which increased with larger administered doses, ranged from 2.2 for low-dose cyclophosphamide to 23.0 for chlorambucil and melphalan and 33.0 for treosulfan in high doses as defined by the researchers. The risk estimates were relative to those of women who received only RT or surgery, no increased leukemia risk being identified in patients who had RT alone. The risk of leukemia was also increased in patients who were treated with the combination of doxorubicin and cisplatin. Kaldor and coworkers concluded that at least one of the two drugs was leukemogenic.[155] The controversies that are engendered by large-scale studies of this kind are reflected in the editorial that accompanied the article and the brisk correspondence that followed. [156] [157]

Travis and associates underscore these issues by their findings regarding the 4402 10-year survivors among the 32,251 women with ovarian cancer on whom they collected data.[108] They found 1296 SMNs rather than the expected 1014 in such a cohort, the cumulative risk at 20 years of follow-up being 18.2% versus the 11.5% expected in the general population. Although leukemias appeared to be related to chemotherapy and cancers in infradiaphragmatic sites appeared to be related to RT, there were also increased risks of tumors in other sites—for example, breast cancer and ocular melanoma—that are not obviously related to the treatments that were used. Travis and associates therefore postulated genetic or other factors that predispose to ovarian cancer as being responsible.[108] In this way, they echo the surmises of Werner-Wasik and colleagues concerning the genetic bases of SMNs that develop in women with cervical cancer.[103]

Fisher and coworkers reviewed the extensive experience that had accumulated in women who were treated for breast cancer following the protocols of the National Surgical Adjuvant Breast and Bowel Project (NSABP).[158] Using data from the SEER registry for comparison, they reported a relative risk of 24.0 for AML among all patients who were treated with surgery and chemotherapy. The risk was found to be 39.3 among patients under 50 years of age compared with a relative risk of 19.9 among those age 50 or older. The relative risks of AML were 2.6 among all women who were treated with surgery only and 10.3 among those who were treated with surgery and RT. These data confirm the leukemogenicity of melphalan, the alkylating agent that was employed in the chemotherapy trials conducted by the NSABP.[155] Curtis and associates reported the relative risks for acute nonlymphoblastic leukemia or myelodysplastic syndrome of 10.0 and 17.4 among unirradiated and irradiated patients with breast cancer, respectively, who were treated with an alkylating agent.[159]

Greene reviewed the evidence concerning the carcinogenicity of cisplatin in animals and humans and provided a concise survey of the oncogenic potential of other chemotherapeutic agents that are in common use in adults with cancer.[153] He pointed out that cisplatin has many of the properties of an oncogene and is carcinogenic in animal systems, where its effects can be reversed by MESNA (2-mercaptoethane-sulfonate). The evidence implicating cisplatin as a leukemogene in humans is found in situations in which it has been used together with etoposide or doxorubicin. It is not clear whether cisplatin is a cofactor in leukemogenesis when used with other drugs or whether these other drugs, rather than cisplatin, are responsible.

AML, usually of French-American-British M5 morphology, which has a characteristic translocation that involves 11q23, has been detected in patients who were given topoisomerase II inhibitors. [160] [161]The anthracyclines, epipodophyllotoxins, and dactinomycin are such inhibitors.[161] Detourmignies and colleagues implicated inhibitors of topoisomerase II in their report of therapy-associated acute promyelocytic leukemia (t-APL).[84] They pointed out that the same translo cation, t(15;17), can be identified in both de novo and treatment-related acute promyelocytic leukemias that develop after therapy with drugs of this class.

van Leeuwen provided an extensive analysis of AML and myelodysplasia that developed after cancer treatments of various kinds and at different ages.[162] In general, the findings authoritatively confirm the observations of others in that chemotherapy was found to be more leukemogenic than was irradiation, and the latent periods for the appearance of nonlymphatic leukemias after therapy with topoisomerase II inhibitors and with alkylating agents tended to be short (<5 years) and long (5 to 10 years), respectively. There has been a recent report of a 23-year interval between treatment of HD and the appearance of secondary erythroleukemia. This was attributed to the alkylating agents that were used as part of the HD therapy.[163] van Leeuwen's detailed analyses by original tumor type, treatments employed, and age factors are worthy of careful reading by students of this problem.[162]

SOLID TUMORS.

Solid tumors are not often associated with anticancer drugs. Greene listed four breast carcinomas among the few solid tumors that appeared in cisplatin-treated women without there being a convincing causative relationship.[153] A dose-effect relationship between alkylating agent treatment and the occurrence of secondary bone sarcomas was identified in children, and there have been several reports of urothelial carcinomas in adults who were treated with cyclophosphamide.[164] Although the data of Fairchild and coworkers[165] and others[166] suggested that cyclophosphamide was not a significant etiologic factor for bladder cancer, Travis and associates reported a relative risk of 4.5 of bladder cancer among patients with NHL who were treated with cyclophosphamide.[167] The relative risk increased with increasing cumulative dose, being 6.3 for cumulative doses of 20 to 49 g and 14.5 for cumulative doses of 50 g or more. Topical nitrogen mustard has been held responsible for the appearance of skin cancers in patients with mycosis fungoides treated in that way.[168]

HODGKIN'S DISEASE.

There is a voluminous literature concerning the leukemias, NHLs, and solid tumors that are encountered in adults with HD, often with conflicting reports. The major focus here will be placed on two recent comprehensive analyses of the solid tumor and leukemia risks. [169] [170]

Dores and colleagues assessed the relative and absolute excess risks of site-specific SMNs in long-term HD survivors.[169] They analyzed data from 32,591 HD patients, including 1111 25-year survivors, and found 1726 solid tumors among those patients. Cancers of the lung, gastrointestinal tract, and female breast were the most frequently observed. The actuarial SMN rate among 25-year survivors was approximately 20%, the risk being about the same in all age groups. Of interest was an apparent decrease in SMN risk after the 25th year of survival. Brusamolino and coworkers reviewed the leukemia risk in HD survivors.[170] Their 1659-patient sample was analyzed according to age, splenectomy, combined modality therapy, and cumulative drug doses, especially of alkylating agents, including nitrosourea derivatives. The overall actuarial risk of leukemia at 15 years was 4.2%, with two peaks. These occurred at 3 and 8 years after initiation of therapy, and the curve flattened at 12 years. The risks after RT alone, chemotherapy alone, and combined modality treatments were 0.3%, 2.8%, and 5.4%, respectively. Risks were higher among patients who had received extended field RT, lomustine, or mechlorethamine. Neither age nor splenectomy proved to be a significant independent variable. No leukemias were found in patients who were treated with ABVD (adriamycin, bleomycin, vinblastine, dacarbazine).

SECONDARY NON-HODGKIN'S LYMPHOMA.

The cumulative incidence rate of secondary NHL among HD patients increased during the first 5 years after treatment before reaching a plateau. These findings were not correlated with any specific treatment or combinations of therapies. Swerdlow and colleagues[171] suggest that immunosuppression could be a contributing factor to the occurrence of secondary NHL in HD patients, most if not all of whom have long been known to be immunologically impaired at diagnosis.[172]

Immunosuppression

Immunosuppressive drugs and the human immunodeficiency virus have been linked to lymphomas and other tumors. [173] [174] Kinlen and associates were among the first to conduct a systematic study of cancers appearing in patients who had been given immunosuppressants.[173] In 1979, they reported an increase in the risk of NHL (especially in the brain) in renal transplant recipients (relative risk: 58.6). A large study of SMNs in patients with either primary or secondary NHL showed bidirectional effects with several potentially virally linked cancer sites, suggesting a role for immune suppression. Bone marrow transplantation entailing iatrogenic immunosuppression has also been linked to SMNs. [175] [176] [177] Curtis and colleagues[175] reported that bone marrow transplant survivors are at risk of developing solid tumors with the passage of time. Their analyses were based on 20,000 bone marrow transplant recipients; many different types of SMNs were encountered, including brain and thyroid tumors. These developed only in patients who had been given brain and total body irradiation. The highest risk was in young children (<10 years of age). In contrast, a smaller cohort of 926 patients with long follow-up showed that recipients over 40 years of age and those with female donors were at higher risk of SMNs, and there was no impact from total body irradiation.[176]

Hosing and associates discussed the risks of myelodysplastic syndrome and acute myelogenous leukemia (AML) after therapy.[178] They studied almost 500 patients with NHL who received high-dose chemotherapy and autologous stem cell transplantation and found 22 patients with myelodysplastic syndrome or AML. The risk was highest among patients who received total body irradiation together with cyclophosphamide and etoposide.

Prolonged immunosuppression is associated with the appearance of skin cancers. Liddington and coworkers reported cutaneous carcinomas (mostly of the squamous cell type) in renal transplant patients and estimated the increase in risk to be approximately 100-fold.[179]

Hormones

Hormones and hormonal manipulations have been held responsible for oncogenesis. Various second tumors have been reported in men with prostatic cancer who were treated with estrogens. These include breast cancers, hepatomas, and desmoids. [180] [181] [182] In women, the use of oral contraceptives that emphasize the estrogenic component has been held responsible for the subsequent appearance of endometrial cancer. Thus, female patients with cancer who are managed by hormonal manipulations or women who use hormone-based contraceptives may be at an increased cancer risk.[183]

Recommendations and Conclusions

The recommendations that were made previously for children apply equally to adults, suitably modified for the age group. In adults particularly, for whom the prognosis often is worse than that for children, the potential risk of a second malignant process should not lead to therapeutic compromises when treatment, often aggressive, is of known benefit. Further SMN research must take into fuller account possible contributory genetic, immunologic, environmental, and other factors beyond which drugs and what RT doses were used.

ACKNOWLEDGMENTS

The authors thank Mrs. Diane Piacente and Mrs. Lee Sucher for preparation of the manuscript.

REFERENCES

  1. Westermeier T, Kaatsch P, Schoetzau A, et al: Multiple primary neoplasms in childhood: data from the German Children's Cancer Registry.  Eur J Cancer1998; 34:687-693.
  2. Green DM, Zevon MA, Reese PA, et al: Second malignant tumors following treatment during childhood and adolescence for cancer.  Med Pediatr Oncol1994; 22:1-10.
  3. Olsen JH, Garwicz S, Hertz H, et al: Second malignant neoplasms after cancer in childhood or adolescence: Nordic Society of Paediatric Haematology and Oncology Association of the Nordic Cancer Registries.  BMJ1993; 307:1030-1036.
  4. Hawkins MM, Draper GJ, Kingston JE: Incidence of second primary tumours among childhood cancer survivors.  Br J Cancer1987; 56:339-347.
  5. de Vathaire F, Hardiman C, Shamsaldin A, et al: Thyroid carcinomas after irradiation for a first cancer during childhood.  Arch Intern Med1999; 159:2713-2719.
  6. Neglia JP, Friedman DL, Yasui Y, et al: Second malignant neoplasms in five-year survivors of childhood cancer: childhood cancer survivor study.  J Natl Cancer Inst2001; 93:618-629.
  7. de Vathaire F, Schweisguth O, Rodary C, et al: Long-term risk of second malignant neoplasm after a cancer in childhood.  Br J Cancer1989; 59:448-452.
  8. Jenkinson HC, Hawkins MM, Stiller CA, et al: Long-term population-based risks of second malignant neoplasms after childhood cancer in Britain.  Br J Cancer2004; 91:1905-1910.
  9. Moller TR, Garwicz S, Barlow L, et al: Decreasing late mortality among five-year survivors of cancer in childhood and adolescence: a population-based study in the Nordic countries.  J Clin Oncol2001; 19:3173-3181.
  10. Lawless SC, Verma P, Green DM, et al: Mortality experiences among 15+ year survivors of childhood and adolescent cancers.  Pediatr Blood Cancer2007; 48:333-338.
  11. Wong FL, Boice Jr JD, Abramson DH, et al: Cancer incidence after retinoblastoma. Radiation dose and sarcoma risk.  JAMA1997; 278:1262-1267.
  12. Kleinerman RA, Tucker MA, Abramson DH, et al: Risk of soft tissue sarcomas by individual subtype in survivors of hereditary retinoblastoma.  J Natl Cancer Inst2007; 99:24-31.
  13. Fletcher O, Easton D, Anderson K, et al: Lifetime risks of common cancers among retinoblastoma survivors.  J Natl Cancer Inst2004; 96:357-363.
  14. Li FP, Fraumeni Jr JF, Mulvihill JJ, et al: A cancer family syndrome in twenty-four kindreds.  Cancer Res1988; 48:5358-5362.
  15. Santibanez-Koref MF, Birch JM, Hartley AL, et al: p53 germline mutations in Li-Fraumeni syndrome.  Lancet1991; 338:1490-1491.
  16. Law JC, Strong LC, Chidambaram A, et al: A germ line mutation in exon 5 of the p53 gene in an extended cancer family.  Cancer Res1991; 51:6385-6387.
  17. Malkin D, Li FP, Strong LC, et al: Germ line p53 mutations in a familial syndrome of breast cancer, sarcomas, and other neoplasms.  Science1990; 250:1233-1238.
  18. Srivastava S, Zou ZQ, Pirollo K, et al: Germline transmission of a mutated p53 gene in a cancer-prone family with Li-Fraumeni syndrome.  Nature1990; 348:747-749.
  19. Malkin D, Jolly KW, Barbier N, et al: Germline mutations of the p53 tumor-suppressor gene in children and young adults with second malignant neoplasms.  N Engl J Med1992; 326:1309-1315.
  20. Malkin D, Jolly KW, Barbier N, et al: Germline mutations of the p53 tumor-suppressor gene in children and young adults with second malignant neoplasms.  N Engl J Med1992; 326:1309-1315.
  21. Garwicz S, Anderson H, Olsen JH, et al: Second malignant neoplasms after cancer in childhood and adolescence: a population-based case-control study in the 5 Nordic countries: The Nordic Society for Pediatric Hematology and Oncology. The Association of the Nordic Cancer Registries.  Int J Cancer2000; 88:672-678.
  22. Sooriyaarachchi GS, Johnson RO, Carbone PP: Neoplasms of the large bowel following ureterosigmoidostomy.  Arch Surg1977; 112:1174-1177.
  23. Recht KA, Belis JA, Kandzari SJ, et al: Ureterosigmoidostomy followed by carcinoma of the colon.  Cancer1979; 44:1538-1542.
  24. Guerin S, Guibout C, Shamsaldin A, et al: Concomitant chemo-radiotherapy and local dose of radiation as risk factors for second malignant neoplasms after solid cancer in childhood: a case-control study.  Int J Cancer2007; 120:96-102.
  25. Haddy N, Le Deley MC, Samand A, et al: Role of radiotherapy and chemotherapy in the risk of secondary leukaemia after a solid tumour in childhood.  Eur J Cancer2006; 42:2757-2764.
  26. Bassal M, Mertens AC, Taylor L, et al: Risk of selected subsequent carcinomas in survivors of childhood cancer: a report from the Childhood Cancer Survivor Study.  J Clin Oncol2006; 24:476-483.
  27. Roggli VL, Estrada R, Fechner RE: Thyroid neoplasia following irradiation for medulloblastoma: report of two cases.  Cancer1979; 43:2232-2238.
  28. Tang TT, Holcenberg JS, Duck SC, et al: Thyroid carcinoma following treatment for acute lymphoblastic leukemia.  Cancer1980; 46:1572-1576.
  29. Moroff SV, Fuks JZ: Thyroid cancer following radiotherapy for Hodgkin's disease: a case report and review of the literature.  Med Pediatr Oncol1986; 14:216-220.
  30. Marus G, Levin CV, Rutherfoord GS: Malignant glioma following radiotherapy for unrelated primary tumors.  Cancer1986; 58:886-894.
  31. Acharya S, Sarafoglou K, LaQuaglia M, et al: Thyroid neoplasms after therapeutic radiation for malignancies during childhood or adolescence.  Cancer2003; 97:2397-2403.
  32. Ron E, Lubin JH, Shore RE, et al: Thyroid cancer after exposure to external radiation: a pooled analysis of seven studies.  Radiat Res1995; 141:259-277.
  33. Upton AC: The dose-response relation in radiation-induced cancer.  Cancer Res1961; 21:717-729.
  34. Gray LH: Cellular Radiation Biology,  Baltimore, Williams and Wilkins, 1965.
  35. Neglia JP, Robison LL, Stovall M, et al: New primary neoplasms of the central nervous system in survivors of childhood cancer: a report from the Childhood Cancer Survivor Study.  J Natl Cancer Inst2006; 98:1528-1537.
  36. Loning L, Zimmermann M, Reiter A, et al: Secondary neoplasms subsequent to Berlin-Frankfurt-Munster therapy of acute lymphoblastic leukemia in childhood: significantly lower risk without cranial radiotherapy.  Blood2000; 95:2770-2775.
  37. Sharif S, Ferner R, Birch JM, et al: Second primary tumors in neurofibromatosis 1 patients treated for optic glioma: substantial risks after radiotherapy.  J Clin Oncol2006; 24:2570-2575.
  38. Rapley EA, Barfoot R, Bonaiti-Pellie C, et al: Evidence for susceptibility genes to familial Wilms tumour in addition to WT1, FWT1 and FWT2.  Br J Cancer2000; 83:177-183.
  39. McDonald JM, Douglass EC, Fisher R, et al: Linkage of familial Wilms' tumor predisposition to chromosome 19 and a two-locus model for the etiology of familial tumors.  Cancer Res1998; 58:1387-1390.
  40. Rahman N, Arbour L, Tonin P, et al: Evidence for a familial Wilms' tumour gene (FWT1) on chromosome 17q12-q21.  Nat Genet1996; 13:461-463.
  41. Pelletier J, Bruening W, Kashtan CE, et al: Germline mutations in the Wilms' tumor suppressor gene are associated with abnormal urogenital development in Denys-Drash syndrome.  Cell1991; 67:437-447.
  42. Bruening W, Bardeesy N, Silverman BL, et al: Germline intronic and exonic mutations in the Wilms' tumour gene (WT1) affecting urogenital development.  Nat Genet1992; 1:144-148.
  43. Li FP, Yan JC, Sallan S, et al: Second neoplasms after Wilms' tumor in childhood.  J Natl Cancer Inst1983; 71:1205-1209.
  44. Breslow NE, Takashima JR, Whitton JA, et al: Second malignant neoplasms following treatment for Wilm's tumor: a report from the National Wilms' Tumor Study Group.  J Clin Oncol1995; 13:1851-1859.
  45. Hartley AL, Birch JM, Blair V, et al: Second primary neoplasms in a population-based series of patients diagnosed with renal tumours in childhood.  Med Pediatr Oncol1994; 22:318-324.
  46. Tucker MA, D'Angio GJ, Boice Jr JD, et al: Bone sarcomas linked to radiotherapy and chemotherapy in children.  N Engl J Med1987; 317:588-593.
  47. Le Vu B, de Vathaire F, Shamsaldin A, et al: Radiation dose, chemotherapy and risk of osteosarcoma after solid tumours during childhood.  Int J Cancer1998; 77:370-377.
  48. Hawkins MM, Wilson LM, Burton HS, et al: Radiotherapy, alkylating agents, and risk of bone cancer after childhood cancer.  J Natl Cancer Inst1996; 88:270-278.
  49. Kuttesch Jr JF, Wexler LH, Marcus RB, et al: Second malignancies after Ewing's sarcoma: radiation dose-dependency of secondary sarcomas.  J Clin Oncol1996; 14:2818-2825.
  50. O'Malley B, D'Angio GJ, Vawter GF: Late effects of roentgen therapy given in infancy.  Am J Roentgenol1963; 89:1067-1073.
  51. Miller AB, How GR, Sherman GJ, et al: Mortality from breast cancer after irradiation during fluoroscopic examinations in patients being treated for tuberculosis.  N Engl J Med1989; 321:1285-1289.
  52. Tokunaga M, Land CE, Yamamoto T, et al: Incidence of female breast cancer among atomic bomb survivors, Hiroshima and Nagasaki, 1950–1980.  Radiat Res1987; 112:243-272.
  53. Taylor AJ, Winter DL, Stiller CA, et al: Risk of breast cancer in female survivors of childhood Hodgkin's disease in Britain: a population-based study.  Int J Cancer2007; 120:384-391.
  54. Bhatia S, Louie AD, Bhatia R, et al: Solid cancers after bone marrow transplantation.  J Clin Oncol2001; 19:464-471.
  55. Ergun-Longmire B, Mertens AC, Mitby P, et al: Growth hormone treatment and risk of second neoplasms in the childhood cancer survivor.  J Clin Endocrinol Metab2006; 91:3494-3498.
  56. Bhatia S, Robison LL, Oberlin O, et al: Breast cancer and other second neoplasms after childhood Hodgkin's disease.  N Engl J Med1996; 334:745-751.
  57. Donaldson SS, Link MP: Combined modality treatment with low-dose radiation and MOPP chemotherapy for children with Hodgkin's disease.  J Clin Oncol1987; 5:742-749.
  58. Schellong G, Riepenhausen M, Creutzig U, et al: Low risk of secondary leukemias after chemotherapy without mechlorethamine in childhood Hodgkin's disease: German-Austrian Pediatric Hodgkin's Disease Group.  J Clin Oncol1997; 15:2247-2253.
  59. de Vathaire F, Francois P, Hill C, et al: Role of radiotherapy and chemotherapy in the risk of second malignant neoplasms after cancer in childhood.  Br J Cancer1989; 59:792-796.
  60. Klein G, Michaelis J, Spix C, et al: Second malignant neoplasms after treatment of childhood cancer.  Eur J Cancer2003; 39:808-817.
  61. Pui CH, Behm FG, Raimondi SC, et al: Secondary acute myeloid leukemia in children treated for acute lymphoid leukemia.  N Engl J Med1989; 321:136-142.
  62. Pui CH, Ribeiro RC, Hancock ML, et al: Acute myeloid leukemia in children treated with epipodophyllotoxins for acute lymphoblastic leukemia.  N Engl J Med1991; 325:1682-1687.
  63. Pedersen-Bjergaard J, Daugaard G, Hansen SW, et al: Increased risk of myelodysplasia and leukaemia after etoposide, cisplatin, and bleomycin for germ-cell tumours.  Lancet1991; 338:359-363.
  64. Winick NJ, McKenna RW, Shuster JJ, et al: Secondary acute myeloid leukemia in children with acute lymphoblastic leukemia treated with etoposide.  J Clin Oncol1993; 11:209-217.
  65. Sugita K, Furukawa T, Tsuchida M, et al: High frequency of etoposide (VP-16)-related secondary leukemia in children with non-Hodgkin's lymphoma.  Am J Pediatr Hematol Oncol1993; 15:99-104.
  66. Hawkins MM, Wilson LM, Stovall MA, et al: Epipodophyllotoxins, alkylating agents, and radiation and risk of secondary leukaemia after childhood cancer.  BMJ1992; 304:951-958.
  67. Tucker MA, Meadows AT, Boice Jr JD, et al: Leukemia after therapy with alkylating agents for childhood cancer.  J Natl Cancer Inst1987; 78:459-464.
  68. Le Deley MC, Leblanc T, Shamsaldin A, et al: Risk of secondary leukemia after a solid tumor in childhood according to the dose of epipodophyllotoxins and anthracyclines: a case-control study by the Societe Francaise d'Oncologie Pediatrique.  J Clin Oncol2003; 21:1074-1081.
  69. Beaumont M, Sanz M, Carli PM, et al: Therapy-related acute promyelocytic leukemia.  J Clin Oncol2003; 21:2123-2137.
  70. Aplan PD, Chervinsky DS, Stanulla M, et al: Site-specific DNA cleavage within the MLL breakpoint cluster region induced by topoisomerase II inhibitors.  Blood1996; 87:2649-2658.
  71. Stanulla M, Wang J, Chervinsky DS, et al: DNA cleavage within the MLL breakpoint cluster region is a specific event which occurs as part of higher-order chromatin fragmentation during the initial stages of apoptosis.  Mol Cell Biol1997; 17:4070-4079.
  72. Stanulla M, Wang J, Chervinsky DS, et al: Topoisomerase II inhibitors induce DNA double-strand breaks at a specific site within the AML1 locus.  Leukemia1997; 11:490-496.
  73. Draper GJ, Sanders BM, Kingston JE: Second primary neoplasms in patients with retinoblastoma.  Br J Cancer1986; 53:661-671.
  74. Bhatia S, Ramsay NK, Steinbuch M, et al: Malignant neoplasms following bone marrow transplantation.  Blood1996; 87:3633-3639.
  75. Mulhern RK, Tyc VL, Phipps S, et al: Health-related behaviors of survivors of childhood cancer.  Med Pediatr Oncol1995; 25:159-165.
  76. Tao ML, Guo MD, Weiss R, et al: Smoking in adult survivors of childhood acute lymphoblastic leukemia.  J Natl Cancer Inst1998; 90:219-225.
  77. Haupt R, Byrne J, Connelly RR, et al: Smoking habits in survivors of childhood and adolescent cancer.  Med Pediatr Oncol1992; 20:301-306.
  78. Travis LB: Therapy-associated solid tumors.  Acta Oncol2002; 41:323-333.
  79. Shen M, Boffetta P, Olsen JH, et al: A pooled analysis of second primary pancreatic cancer.  Am J Epidemiol2006; 163:502-511.
  80. Hemminki K, Jiang Y, Steineck G: Skin cancer and non-Hodgkin's lymphoma as second malignancies.  Markers of impaired immune function? Eur J Cancer2003; 39:223-229.
  81. Swift M, Morrell D, Massey RB, et al: Incidence of cancer in 161 families affected by ataxia-telangiectasia.  N Engl J Med1991; 325:1831-1836.
  82. Little MP, Muirhead CR, Haylock RG, et al: Relative risks of radiation-associated cancer: comparison of second cancer in therapeutically irradiated populations with the Japanese atomic bomb survivors.  Radiat Environ Biophys1999; 38:267-283.
  83. Le Beau MM, Albain KS, Larson RA, et al: Clinical and cytogenetic correlations in 63 patients with therapy-related myelodysplastic syndromes and acute nonlymphocytic leukemia: further evidence for characteristic abnormalities of chromosomes no. 5 and 7.  J Clin Oncol1986; 4:325-345.
  84. Detourmignies L, Castaigne S, Stoppa AM, et al: Therapy-related acute promyelocytic leukemia: a report on 16 cases.  J Clin Oncol1992; 10:1430-1435.
  85. Collins VP, Nordenskjold M, Dumanski JP: The molecular genetics of meningiomas.  Brain Pathol1990; 1:19-24.
  86. el-Azouzi M, Chung RY, Farmer GE, et al: Loss of distinct regions on the short arm of chromosome 17 associated with tumorigenesis of human astrocytomas.  Proc Natl Acad Sci USA1989; 86:7186-7190.
  87. Schiffman S, Berger A: Stewart-Treves syndrome.  J Am Coll Surg2007; 204:328.
  88. Edeiken S, Russo DP, Knecht J, et al: Angiosarcoma after tylectomy and radiation therapy for carcinoma of the breast.  Cancer1992; 70:644-647.
  89. Marchal C, Weber B, de Lafontan B, et al: Nine breast angiosarcomas after conservative treatment for breast carcinoma: a survey from French comprehensive Cancer Centers.  Int J Radiat Oncol Biol Phys1999; 44:113-119.
  90. Lindsay KA, Wheldon EG, Deehan C, et al: Radiation carcinogenesis modelling for risk of treatment-related second tumours following radiotherapy.  Br J Radiol2001; 74:529-536.
  91. Sachs RK, Brenner DJ: Solid tumor risks after high doses of ionizing radiation.  Proc Natl Acad Sci USA2005; 102:13040-13045.
  92. Behringer K, Josting A, Schiller P, et al: Solid tumors in patients treated for Hodgkin's disease: a report from the German Hodgkin Lymphoma Study Group.  Ann Oncol2004; 15:1079-1085.
  93. Travis LB, Hill D, Dores GM, et al: Cumulative absolute breast cancer risk for young women treated for Hodgkin lymphoma.  J Natl Cancer Inst2005; 97:1428-1437.
  94. Wendland MM, Tsodikov A, Glenn MJ, et al: Time interval to the development of breast carcinoma after treatment for Hodgkin disease.  Cancer2004; 101:1275-1282.
  95. Tward JD, Wendland MM, Shrieve DC, et al: The risk of secondary malignancies over 30 years after the treatment of non-Hodgkin lymphoma.  Cancer2006; 107:108-115.
  96. Ji J, Hemminki K: Second primary malignancies among patients with soft tissue tumors in Sweden.  Int J Cancer2006; 119:909-914.
  97. Lagrange JL, Ramaioli A, Chateau MC, et al: Sarcoma after radiation therapy: retrospective multiinstitutional study of 80 histologically confirmed cases: Radiation Therapist and Pathologist Groups of the Federation Nationale des Centres de Lutte Contre le Cancer.  Radiology2000; 216:197-205.
  98. Tountas AA, Fornasier VL, Harwood AR, Leung PM: Postirradiation sarcoma of bone: a perspective.  Cancer1979; 43:182-187.
  99. Dikshit RP, Boffetta P, Bouchardy C, et al: Risk factors for the development of second primary tumors among men after laryngeal and hypopharyngeal carcinoma.  Cancer2005; 103:2326-2333.
  100. Bairati I, Meyer F, Gelinas M, et al: A randomized trial of antioxidant vitamins to prevent second primary cancers in head and neck cancer patients.  J Natl Cancer Inst2005; 97:481-488.
  101. Kong L, Lu JJ, Hu C, et al: The risk of second primary tumors in patients with nasopharyngeal carcinoma after definitive radiotherapy.  Cancer2006; 107:1287-1293.
  102. Storm HH: Second primary cancer after treatment for cervical cancer: late effects after radiotherapy.  Cancer1988; 61:679-688.
  103. Werner-Wasik M, Schmid CH, Bornstein LE, et al: Increased risk of second malignant neoplasms outside radiation fields in patients with cervical carcinoma.  Cancer1995; 75:2281-2285.
  104. Sturgeon SR, Curtis RE, Johnson K, et al: Second primary cancers after vulvar and vaginal cancers.  Am J Obstet Gynecol1996; 174:929-933.
  105. Hemminki K, Dong C: Cancer in husbands of cervical cancer patients.  Epidemiology2000; 11:347-349.
  106. Hall HI, Jamison P, Weir HK: Second primary ovarian cancer among women diagnosed previously with cancer.  Cancer Epidemiol Biomarkers Prev2001; 10:995-999.
  107. Bergfeldt K, Nilsson B, Einhorn S, et al: Breast cancer risk in women with a primary ovarian cancer: a case-control study.  Eur J Cancer2001; 37:2229-2234.
  108. Travis LB, Curtis RE, Boice Jr JD, et al: Second malignant neoplasms among long-term survivors of ovarian cancer.  Cancer Res1996; 56:1564-1570.
  109. Buiatti E, Crocetti E, Acciai S, et al: Incidence of second primary cancers in three Italian population-based cancer registries.  Eur J Cancer1997; 33:1829-1834.
  110. Movsas B, Hanlon AL, Pinover W, et al: Is there an increased risk of second primaries following prostate irradiation?.  Int J Radiat Oncol Biol Phys1998; 41:251-255.
  111. Johnstone PA, Powell CR, Riffenburgh R, et al: The fate of 10-year clinically recurrence-free survivors after definitive radiotherapy for T1-3N0M0 prostate cancer.  Radiat Oncol Investig1998; 6:103-108.
  112. Baxter NN, Tepper JE, Durham SB, et al: Increased risk of rectal cancer after prostate radiation: a population-based study.  Gastroenterology2005; 128:819-824.
  113. Brenner DJ, Curtis RE, Hall EJ, et al: Second malignancies in prostate carcinoma patients after radiotherapy compared with surgery.  Cancer2000; 88:398-406.
  114. Kendal WS, Eapen L, Macrae R, et al: Prostatic irradiation is not associated with any measurable increase in the risk of subsequent rectal cancer.  Int J Radiat Oncol Biol Phys2006; 65:661-668.
  115. Moon K, Stukenborg GJ, Keim J, et al: Cancer incidence after localized therapy for prostate cancer.  Cancer2006; 107:991-998.
  116. Kendal W, Eapen L, Nicholas G: Second primary cancers after prostatic irradiation: ensuring an appropriate analysis.  Cancer2007; 109:164.author reply 165
  117. Wanderas EH, Fossa SD, Tretli S: Risk of a second germ cell cancer after treatment of a primary germ cell cancer in 2201 Norwegian male patients.  Eur J Cancer1997; 33:244-252.
  118. Ruther U, Dieckmann KP, Bussar-Maatz R, et al: Second malignancies following pure seminoma.  Oncology2000; 58:75-82.
  119. Moller H, Mellemgaard A, Jacobsen GK, et al: Incidence of second primary cancer following testicular cancer!.  Eur J Cancer1993; 29A:672-676.
  120. Travis LB, Fossa SD, Schonfeld SJ, et al: Second cancers among 40,576 testicular cancer patients: focus on long-term survivors.  J Natl Cancer Inst2005; 97:1354-1365.
  121. Richiardi L, Scelo G, Boffetta P, et al: Second malignancies among survivors of germ-cell testicular cancer: a pooled analysis between 13 cancer registries.  Int J Cancer2007; 120:623-631.
  122. Nichols CR, Loehrer Sr PJ: The story of second cancers in patients cured of testicular cancer: tarnishing success or burnishing irrelevance?.  J Natl Cancer Inst1997; 89:1394-1395.
  123. Raymond JS, Hogue CJ: Multiple primary tumours in women following breast cancer, 1973–2000.  Br J Cancer2006; 94:1745-1750.
  124. Metcalfe K, Lynch HT, Ghadirian P, et al: Contralateral breast cancer in BRCA1 and BRCA2 mutation carriers.  J Clin Oncol2004; 22:2328-2335.
  125. Satram-Hoang S, Ziogas A, Anton-Culver H: Risk of second primary cancer in men with breast cancer.  Breast Cancer Res2007; 9:R10.
  126. Hemminki K, Scelo G, Boffetta P, et al: Second primary malignancies in patients with male breast cancer.  Br J Cancer2005; 92:1288-1292.
  127. Fowble B, Hanlon A, Freedman G, et al: Second cancers after conservative surgery and radiation for stages I–II breast cancer: identifying a subset of women at increased risk.  Int J Radiat Oncol Biol Phys2001; 51:679-690.
  128. Hill-Kayser CE, Harris EE, Hwang WT, et al: Twenty-year incidence and patterns of contralateral breast cancer after breast conservation treatment with radiation.  Int J Radiat Oncol Biol Phys2006; 66:1313-1319.
  129. Fraass BA, Roberson PL, Lichter AS: Dose to the contralateral breast due to primary breast irradiation.  Int J Radiat Oncol Biol Phys1985; 11:485-497.
  130. Boice Jr JD, Harvey EB, Blettner M, et al: Cancer in the contralateral breast after radiotherapy for breast cancer.  N Engl J Med1992; 326:781-785.
  131. Yu GP, Schantz SP, Neugut AI, et al: Incidences and trends of second cancers in female breast cancer patients: a fixed inception cohort-based analysis (United States).  Cancer Causes Control2006; 17:411-420.
  132. Levi F, Randimbison L, Te VC, et al: Cancer risk after radiotherapy for breast cancer.  Br J Cancer2006; 95:390-392.
  133. Clarke M, Collins R, Darby S, et al: Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials.  Lancet2005; 366:2087-2106.
  134. Roychoudhuri R, Evans H, Robinson D, et al: Radiation-induced malignancies following radiotherapy for breast cancer.  Br J Cancer2004; 91:868-872.
  135. Obedian E, Fischer DB, Haffty BG: Second malignancies after treatment of early-stage breast cancer: lumpectomy and radiation therapy versus mastectomy.  J Clin Oncol2000; 18:2406-2412.
  136. Levi F, Randimbison L, Te VC, et al: Increased risk of esophageal cancer after breast cancer.  Ann Oncol2005; 16:1829-1831.
  137. Zablotska LB, Chak A, Das A, et al: Increased risk of squamous cell esophageal cancer after adjuvant radiation therapy for primary breast cancer.  Am J Epidemiol2005; 161:330-337.
  138. Jones A: Radiation oncogenesis in relation to the treatment of pituitary tumours.  Clin Endocrinol (Oxf)1991; 35:379-397.
  139. Erfurth EM, Bulow B, Mikoczy Z, et al: Is there an increase in second brain tumours after surgery and irradiation for a pituitary tumour?.  Clin Endocrinol (Oxf)2001; 55:613-616.
  140. Schonfeld SJ, Gilbert ES, Dores GM, et al: Acute myeloid leukemia following Hodgkin lymphoma: a population-based study of 35,511 patients.  J Natl Cancer Inst2006; 98:215-218.
  141. Kry SF, Salehpour M, Followill DS, et al: The calculated risk of fatal secondary malignancies from intensity-modulated radiation therapy.  Int J Radiat Oncol Biol Phys2005; 62:1195-1203.
  142. Hall EJ: Intensity-modulated radiation therapy, protons, and the risk of second cancers.  Int J Radiat Oncol Biol Phys2006; 65:1-7.
  143. Gottschalk B: Neutron dose in scattered and scanned proton beams: in regard to Eric J. Hall (Int J Radiat Oncol Biol Phys 2006;65:1–7).  Int J Radiat Oncol Biol Phys2006; 66:1594.author reply 1595
  144. Hall EJ, Brenner DJ: In reply to Drs. Macklis, Gottschalk, Paganetti, et al.  Int J Radiat Oncol Biol Phys2006; 66:1595.
  145. Rubino C, de Vathaire F, Dottorini ME, et al: Second primary malignancies in thyroid cancer patients.  Br J Cancer2003; 89:1638-1644.
  146. Sandeep TC, Strachan MW, Reynolds RM, et al: Second primary cancers in thyroid cancer patients: a multinational record linkage study.  J Clin Endocrinol Metab2006; 91:1819-1825.
  147. Ronckers CM, McCarron P, Ron E: Thyroid cancer and multiple primary tumors in the SEER cancer registries.  Int J Cancer2005; 117:281-288.
  148. Verkooijen RB, Smit JW, Romijn JA, et al: The incidence of second primary tumors in thyroid cancer patients is increased, but not related to treatment of thyroid cancer.  Eur J Endocrinol2006; 155:801-806.
  149. Kyle RA, Pierre RV, Bayrd ED: Multiple myeloma and acute myelomonocytic leukemia.  N Engl J Med1970; 283:1121-1125.
  150. Reimer RR, Hoover R, Fraumeni Jr JF, et al: Acute leukemia after alkylating-agent therapy of ovarian cancer.  N Engl J Med1977; 297:177-181.
  151. Boice Jr JD, Greene MH, Killen Jr JY, et al: Leukemia and preleukemia after adjuvant treatment of gastrointestinal cancer with semustine (methyl-CCNU).  N Engl J Med1983; 309:1079-1084.
  152. Greene MH, Boice Jr JD, Strike TA: Carmustine as a cause of acute nonlymphocytic leukemia.  N Engl J Med1985; 313:579.
  153. Greene MH: Is cisplatin a human carcinogen?.  J Natl Cancer Inst1992; 84:306-312.
  154. Greene MH, Boice Jr JD, Greer BE, et al: Acute nonlymphocytic leukemia after therapy with alkylating agents for ovarian cancer: a study of five randomized clinical trials.  N Engl J Med1982; 307:1416-1421.
  155. Kaldor JM, Day NE, Pettersson F, et al: Leukemia following chemotherapy for ovarian cancer.  N Engl J Med1990; 322:1-6.
  156. Leukemia after treatment of ovarian cancer or Hodgkin's disease.  N Engl J Med1990; 322:1818-1820.
  157. Coltman Jr CA, Dahlberg S: Treatment-related leukemia.  N Engl J Med1990; 322:52-53.
  158. Fisher B, Rockette H, Fisher ER, et al: Leukemia in breast cancer patients following adjuvant chemotherapy or postoperative radiation: the NSABP experience.  J Clin Oncol1985; 3:1640-1658.
  159. Curtis RE, Boice Jr JD, Stovall M, et al: Risk of leukemia after chemotherapy and radiation treatment for breast cancer.  N Engl J Med1992; 326:1745-1751.
  160. Pedersen-Bjergaard J, Sigsgaard TC, Nielsen D, et al: Acute monocytic or myelomonocytic leukemia with balanced chromosome translocations to band 11q23 after therapy with 4-epi-doxorubicin and cisplatin or cyclophosphamide for breast cancer.  J Clin Oncol1992; 10:1444-1451.
  161. Smith MA, Rubinstein L, Ungerleider RS: Therapy-related acute myeloid leukemia following treatment with epipodophyllotoxins: estimating the risks.  Med Pediatr Oncol1994; 23:86-98.
  162. van Leeuwen FE: Risk of acute myelogenous leukaemia and myelodysplasia following cancer treatment.  Baillieres Clin Haematol1996; 9:57-85.
  163. Scully RE, Mark EJ, McNeely WF, et al: Case records of the Massachusetts General Hospital. Weekly clinicopathological exercises. Case 20-1997. A 74-year-old man with progressive cough, dyspnea, and pleural thickening.  N Engl J Med1997; 336:1895-1903.
  164. Durkee C, Benson Jr R: Bladder cancer following administration of cyclophosphamide.  Urology1980; 16:145-148.
  165. Fairchild WV, Spence CR, Solomon HD, et al: The incidence of bladder cancer after cyclophosphamide therapy.  J Urol1979; 122:163-164.
  166. Pearson RM, Soloway MS: Does cyclophosphamide induce bladder cancer?.  Urology1978; 11:437-447.
  167. Travis LB, Curtis RE, Glimelius B, et al: Bladder and kidney cancer following cyclophosphamide therapy for non-Hodgkin's lymphoma.  J Natl Cancer Inst1995; 87:524-530.
  168. Lee LA, Fritz KA, Golitz L, et al: Second cutaneous malignancies in patients with mycosis fungoides treated with topical nitrogen mustard.  J Am Acad Dermatol1982; 7:590-598.
  169. Dores GM, Metayer C, Curtis RE, et al: Second malignant neoplasms among long-term survivors of Hodgkin's disease: a population-based evaluation over 25 years.  J Clin Oncol2002; 20:3484-3494.
  170. Brusamolino E, Anselmo AP, Klersy C, et al: The risk of acute leukemia in patients treated for Hodgkin's disease is significantly higher after combined modality programs than after chemotherapy alone and is correlated with the extent of radiotherapy and type and duration of chemotherapy: a case-control study.  Haematologica1998; 83:812-823.
  171. Swerdlow AJ, Barber JA, Horwich A, et al: Second malignancy in patients with Hodgkin's disease treated at the Royal Marsden Hospital.  Br J Cancer1997; 75:116-123.
  172. van Rijswijk RE, Sybesma JP, Kater L: A prospective study of the changes in immune status following radiotherapy for Hodgkin's disease.  Cancer1984; 53:62-69.
  173. Kinlen LJ, Sheil AG, Peto J, et al: Collaborative United Kingdom-Australasian study of cancer in patients treated with immunosuppressive drugs.  Br Med J1979; 2:1461-1466.
  174. Snider WD, Simpson DM, Aronyk KE, Nielsen SL: Primary lymphoma of the nervous system associated with acquired immune-deficiency syndrome.  N Engl J Med1983; 308:45.
  175. Curtis RE, Rowlings PA, Deeg HJ, et al: Solid cancers after bone marrow transplantation.  N Engl J Med1997; 336:897-904.
  176. Gallagher G, Forrest DL: Second solid cancers after allogeneic hematopoietic stem cell transplantation.  Cancer2007; 109:84-92.
  177. Baker KS, DeFor TE, Burns LJ, et al: New malignancies after blood or marrow stem-cell transplantation in children and adults: incidence and risk factors.  J Clin Oncol2003; 21:1352-1358.
  178. Hosing C, Munsell M, Yazji S, et al: Risk of therapy-related myelodysplastic syndrome/acute leukemia following high-dose therapy and autologous bone marrow transplantation for non-Hodgkin's lymphoma.  Ann Oncol2002; 13:450-459.
  179. Liddington M, Richardson AJ, Higgins RM, et al: Skin cancer in renal transplant recipients.  Br J Surg1989; 76:1002-1005.
  180. Dore B, Dombriz M, Denis P, et al: [Cancer of the breast in patients having prostatic cancer treated with estrogens. Report of three cases (author's transl)].  J Urol (Paris)1982; 88:247-252.
  181. Brooks JJ: Hepatoma associated with diethylstilbestrol therapy for prostatic carcinoma.  J Urol1982; 128:1044-1045.
  182. Svanvik J, Knutsson F, Jansson R, et al: Desmoid tumor in the abdominal wall after treatment with high dose estradiol for prostatic cancer.  Acta Chir Scand1982; 148:301-303.
  183. Weiss NS, Sayvetz TA: Incidence of endometrial cancer in relation to the use of oral contraceptives.  N Engl J Med1980; 302:551-554.