A multicenter case-control study. Cancer Epidemiol, Biomarkers & Prev 1:35-43, 1991 (5) JANERICH DT, THOMPSON WD, VARELA LR,
ET AL: Lung cancer and exposure to tobacco smoke in the household. N Engl J Med 323:632-636, 1990 (6) CORREA P, PICKLE LW, FONTHAM E, ET AL:
Passive smoking and lung cancer. Lancet 2:595-597, 1983 (7) SANDLER DP, WILCOX AJ,
EVERSON
RB:
Cumulative effects of lifetime passive smoking on cancer risk. Lancet 1:312-315, 1985 (8) Wu AH, HENDERSON BE, PIKE MC, ET AL:
Smoking and other risk factors for lung cancer in women. JNCI 74:747-751, 1985 (9) DALAGER NA,
PICKLE LW,
MASON TJ, ET
AL: The relation of passive smoking to lung cancer. Cancer Res 46:4808-^t811, 1986 (70)
GARPINKEL L, AUERBACH O, JOUBERT L: In-
voluntary smoking and lung cancer: A casecontrol study. JNCI 75:465-469, 1985 (//)
TRICHOPOULOS D, KALANDIDI L, SPARROS
L, ET AL: Lung cancer and passive smoking. Int J Cancer 27:1-4, 1981 Passive smoking and lung cancer in Swedish women. Am J Epidemiol 125:17-24, 1987 (13) Koo LC, Ho JH, SAW D, ET AL: Measurements of passive smoking and estimates of lung cancer risk among non-smoking Chinese females. Int J Cancer 39:162-169, 1987 (14) LAM TH,
KUNG IT, WONG CM,
ET AL:
Smoking, passive smoking and histological types in lung cancer in Hong Kong Chinese women. Br J Cancer 56:673-678, 1987
Second Cancers in Patients With Chronic Lymphocytic Leukemia Lois B. Travis,* Rochelle E. Curtis, Benjamin F. Hankey, Joseph F. Fraumeni, Jr.
Background: Reports to date have provided widely divergent estimates of the risk of second malignant neoplasms in patients with chronic lymPatients with chronic lymphocytic phocytic leukemia (CLL), ranging leukemia (CLL) exhibit a variety of imfrom cancer deficits to excesses of munologic perturbations {1-4) that may twofold to threefold. Purpose: Our increase their risk for second malignant purpose was to estimate the risk of neoplasms. The occurrence of familial second primary cancers following CLL may also suggest, for some subCLL, utilizing population-based tu- jects, genetic determinants (5-7), such mor registries, and to determine as those that underlie other sets of mulwhether site-specific excesses might tiple primary cancers (8,9). Moreover, be associated with type of initial radiotherapy and chemotherapeutic treatment for CLL. Methods: We agents may also contribute to subse1422
quent malignancies among cancer survivors (10). It is important to clarify the risk of second cancers in CLL patients because of the potential impact on patient management, follow-up, and survival. However, various reports to date have provided divergent estimates of the occurrence of second malignancies in CLL patients, ranging from cancer deficits to excesses of twofold to threefold (11-20). To further explore and quantify the risk of second cancers among a large number of CLL patients in the general population and to examine associations of risk with initial therapy, we conducted a survey of more than 9000 such subjects reported to the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program1 from 1973 through 1988. Since CLL patients are frequently treated only with alkylating agents without the confounding effects of radiotherapy, this group of patients provides a special opportunity to study the late sequelae of these drugs.
Patients and Methods We analyzed all patients diagnosed with CLL as a first primary cancer between 1973 and 1988 who were reported to one of nine population-based cancer registries of the SEER Program and survived 2 or more months. Such registries include those in the metropolitan areas of
Received January 10, 1992; revised May 14, 1992; accepted June 4, 1992. L. B. Travis, R. E. Curtis, J. F. Fraumeni, Jr. (Epidemiology and Biostatistics Program, Division of Cancer Etiology), B. F. Hankey (Cancer Statistics Branch, Surveillance Program, Division of Cancer Prevention and Control), National Cancer Institute, Bethesda, Md. We are indebted to the State Health Registry of Iowa (Ms. Kathleen McKeen and Dr. Charles F. Lynch) for data retrieval, to Drs. John Boice, Neil Caporaso, and Martha Linet for critical review of the manuscript, and to Ms. Sandy Coopersmith and Ms. Shirley Carson for secretarial support. * Correspondence to: Lois B. Travis, M.D., Executive Plaza North, Ste. 408, National Institutes of Health, Bethesda, MD 20892. 1 Ed. note: The program is a set of geographically defined, population-based central tumor registries in the United States, operated by local nonprofit organizations under contract to the National Cancer Institute (NCI). Each registry annually submits its cases to the NCI on a computer tape. These computer tapes are then edited by the NCI and made available for analysis.
Journal of the National Cancer Institute
Downloaded from http://jnci.oxfordjournals.org/ at Brunel University on January 3, 2015
(12) PERSHAGAN G, HRUBEC Z, SVENSSON C:
analyzed data for 9456 patients diagnosed with CLL as a first primary cancer between 1973 and 1988, who were reported to one of nine tumor registries participating in the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program and who survived 2 or more months. SEER files were searched for invasive primary malignancies that developed at least 2 months after the initial CLL diagnosis. Results: Compared with the general population, CLL patients demonstrated a significantly increased risk of developing all second cancers (840 observed; observed-toexpected ratio [O/E] = 1.28; 95% confidence interval [CI] = 1.19-1.37). Significant excesses were noted for cancers of the lung (O/E = 1.90), brain (O/E = 1.98), and eye (intraocular melanoma) (O/E = 3.97) as well as malignant melanoma (O/E = 2.79) and Hodgkin's disease (O/E = 7.69). Cancer risk, which did not vary according to initial treatment category, was also constant across all time intervals after CLL diagnosis. Conclusion: CLL patients are at a significantly increased risk of developing a second malignant neoplasm. The pattern of cancer excesses suggests a susceptibility state permitting the development of selected second malignancies in patients with CLL, perhaps because of shared etiologic factors, immunologic impairment, and/or other influences. Although our results do not suggest a strong treatment effect, more detailed studies of second tumors in CLL are needed to investigate the role of radiation therapy and chemotherapy. [J Natl Cancer Inst 84:1422-1427, 1992]
2
However, the number of patients in whom this sequence of diagnoses was reported is noted.
Vol. 84, No. 18, September 16, 1992
To calculate the risk of second cancers, we assembled person-years of observation according to age, sex, and calendar periods from 2 months after the date of CLL diagnosis to the date of last follow-up evaluation, date of death, date of diagnosis of a second cancer, or the end of the study (December 31, 1988), whichever occurred first. SEER incidence rates specific for 5-year age, sex, and 5-year calendar year intervals were multiplied by the accumulated person-years at risk to estimate the number of cancer cases expected. Statistical tests and 95% confidence limits were based on the assumption that the observed numbers of second tumors were distributed as a Poisson variable. The methods of Breslow et al. (29) were used to conduct tests for homogeneity and linear trend.
Results During the study period, 9456 twomonth survivors of CLL were reported to the SEER registries (Table 1). The average age at diagnosis was 70.0 years, and the average follow-up was 4.22 years. Approximately 92% of the patients were White, and 6.1% were Black. Second cancers developed in 8.9% of subjects. Diagnoses were histo-
logically or cytologically confirmed in 94.6% of the CLL cases and in 94.1% of the second tumors. Initial treatment is presented in Table 2. A second malignancy developed in 840 subjects compared with 658 expected (observed-to-expected ratio [O/E] = 1.28; 95% confidence interval [CI] = 1.19-1.37) (Table 3). Elevated risks were observed for cancers of the lung, eye, and brain and for malignant melanoma and Hodgkin's disease. All ocular tumors were melanomas. A significant excess of prostate cancer was limited to the 1st year after diagnosis of CLL, a pattern consistent with a surveillance effect. Nonsignificant elevations in cancers of the larynx, testis, kidney, and connective tissue were also apparent. The histology of the connective tissue tumors, all of which occurred in men, included malignant fibrous histiocytoma (n = 2), Kaposi's sarcoma (n = 1), and liposarcoma (n = 1). The median time for the development of connective tissue tumors following CLL was 33 months. Malignancies of the gastrointestinal tract were not significantly increased. Decreased risks for cancers of the breast, uterine cervix, and uterine corpus were apparent, although ovarian neoplasms occurred slightly in excess.
Downloaded from http://jnci.oxfordjournals.org/ at Brunel University on January 3, 2015
Detroit, Atlanta (1975-1988), SeattlePuget Sound (1974-1988), and San Francisco-Oakland as well as the states of Connecticut, Hawaii, Iowa, New Mexico, and Utah—areas which comprise approximately 10% of the U.S. population. Patient demographic data, vital status, and tumor histology are routinely reported to the SEER Program. In addition, SEER files also include the type of initial treatment given to patients according to one of several broad designations. For analyses, we grouped patients into the following categories: radiation therapy only, chemotherapy only, radiation therapy and chemotherapy, no treatment, and other treatment. The last category consisted of subjects whose initial therapy was either hormonal (including corticosteroids), immunologic, surgical, or unknown. The SEER Program does not record therapy given after the first course of treatment. In addition, information is not available with regard to the specific drugs administered or dose schedules used. However, patients with CLL are usually treated with continuous or pulse administration of low-dose chlorambucil with or without prednisone (21-23). Less frequently, either cyclophosphamide (24) or combination chemotherapy regimens (25) are used. Radiation therapy is generally used only to treat areas of bulky disease that do not respond to chemotherapy; splenic irradiation is given in selected patients (26,27). SEER files were examined for invasive primary malignancies that developed at least 2 months after the initial CLL diagnosis. Cancer sites were grouped according to the World Health Organization's International Classification of Diseases for Oncology (ICD-O) (28). Because reporting of second diagnoses of leukemia following CLL was incomplete during the early years of the SEER Program, such malignancies were not included in any evaluations. Also, since CLL may represent the leukemic phase of certain types of non-Hodgkin's lymphoma, the latter diagnosis was excluded from analyses of second cancer risk.2
Table 1. Characteristics of patients with CLL reported to the SEER cancer registries, 1973-1988 Category No. with CLL' No. who developed a second primary cancer Average age at diagnosis of CLL, y Person-years of follow-up Average follow-up, y
Men
Women
Total
5508 575 68.5 22 103 4.01
3948 265 72.0 17 840 4.51
9456 840 70.0 39943 4.22
* ICD-0 morphology code 9823. Number excludes all persons who survived
ET AL: Lung cancer and exposure to tobacco smoke in the household. N Engl J Med 323:632-636, 1990 (6) CORREA P, PICKLE LW, FONTHAM E, ET AL:
Passive smoking and lung cancer. Lancet 2:595-597, 1983 (7) SANDLER DP, WILCOX AJ,
EVERSON
RB:
Cumulative effects of lifetime passive smoking on cancer risk. Lancet 1:312-315, 1985 (8) Wu AH, HENDERSON BE, PIKE MC, ET AL:
Smoking and other risk factors for lung cancer in women. JNCI 74:747-751, 1985 (9) DALAGER NA,
PICKLE LW,
MASON TJ, ET
AL: The relation of passive smoking to lung cancer. Cancer Res 46:4808-^t811, 1986 (70)
GARPINKEL L, AUERBACH O, JOUBERT L: In-
voluntary smoking and lung cancer: A casecontrol study. JNCI 75:465-469, 1985 (//)
TRICHOPOULOS D, KALANDIDI L, SPARROS
L, ET AL: Lung cancer and passive smoking. Int J Cancer 27:1-4, 1981 Passive smoking and lung cancer in Swedish women. Am J Epidemiol 125:17-24, 1987 (13) Koo LC, Ho JH, SAW D, ET AL: Measurements of passive smoking and estimates of lung cancer risk among non-smoking Chinese females. Int J Cancer 39:162-169, 1987 (14) LAM TH,
KUNG IT, WONG CM,
ET AL:
Smoking, passive smoking and histological types in lung cancer in Hong Kong Chinese women. Br J Cancer 56:673-678, 1987
Second Cancers in Patients With Chronic Lymphocytic Leukemia Lois B. Travis,* Rochelle E. Curtis, Benjamin F. Hankey, Joseph F. Fraumeni, Jr.
Background: Reports to date have provided widely divergent estimates of the risk of second malignant neoplasms in patients with chronic lymPatients with chronic lymphocytic phocytic leukemia (CLL), ranging leukemia (CLL) exhibit a variety of imfrom cancer deficits to excesses of munologic perturbations {1-4) that may twofold to threefold. Purpose: Our increase their risk for second malignant purpose was to estimate the risk of neoplasms. The occurrence of familial second primary cancers following CLL may also suggest, for some subCLL, utilizing population-based tu- jects, genetic determinants (5-7), such mor registries, and to determine as those that underlie other sets of mulwhether site-specific excesses might tiple primary cancers (8,9). Moreover, be associated with type of initial radiotherapy and chemotherapeutic treatment for CLL. Methods: We agents may also contribute to subse1422
quent malignancies among cancer survivors (10). It is important to clarify the risk of second cancers in CLL patients because of the potential impact on patient management, follow-up, and survival. However, various reports to date have provided divergent estimates of the occurrence of second malignancies in CLL patients, ranging from cancer deficits to excesses of twofold to threefold (11-20). To further explore and quantify the risk of second cancers among a large number of CLL patients in the general population and to examine associations of risk with initial therapy, we conducted a survey of more than 9000 such subjects reported to the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program1 from 1973 through 1988. Since CLL patients are frequently treated only with alkylating agents without the confounding effects of radiotherapy, this group of patients provides a special opportunity to study the late sequelae of these drugs.
Patients and Methods We analyzed all patients diagnosed with CLL as a first primary cancer between 1973 and 1988 who were reported to one of nine population-based cancer registries of the SEER Program and survived 2 or more months. Such registries include those in the metropolitan areas of
Received January 10, 1992; revised May 14, 1992; accepted June 4, 1992. L. B. Travis, R. E. Curtis, J. F. Fraumeni, Jr. (Epidemiology and Biostatistics Program, Division of Cancer Etiology), B. F. Hankey (Cancer Statistics Branch, Surveillance Program, Division of Cancer Prevention and Control), National Cancer Institute, Bethesda, Md. We are indebted to the State Health Registry of Iowa (Ms. Kathleen McKeen and Dr. Charles F. Lynch) for data retrieval, to Drs. John Boice, Neil Caporaso, and Martha Linet for critical review of the manuscript, and to Ms. Sandy Coopersmith and Ms. Shirley Carson for secretarial support. * Correspondence to: Lois B. Travis, M.D., Executive Plaza North, Ste. 408, National Institutes of Health, Bethesda, MD 20892. 1 Ed. note: The program is a set of geographically defined, population-based central tumor registries in the United States, operated by local nonprofit organizations under contract to the National Cancer Institute (NCI). Each registry annually submits its cases to the NCI on a computer tape. These computer tapes are then edited by the NCI and made available for analysis.
Journal of the National Cancer Institute
Downloaded from http://jnci.oxfordjournals.org/ at Brunel University on January 3, 2015
(12) PERSHAGAN G, HRUBEC Z, SVENSSON C:
analyzed data for 9456 patients diagnosed with CLL as a first primary cancer between 1973 and 1988, who were reported to one of nine tumor registries participating in the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program and who survived 2 or more months. SEER files were searched for invasive primary malignancies that developed at least 2 months after the initial CLL diagnosis. Results: Compared with the general population, CLL patients demonstrated a significantly increased risk of developing all second cancers (840 observed; observed-toexpected ratio [O/E] = 1.28; 95% confidence interval [CI] = 1.19-1.37). Significant excesses were noted for cancers of the lung (O/E = 1.90), brain (O/E = 1.98), and eye (intraocular melanoma) (O/E = 3.97) as well as malignant melanoma (O/E = 2.79) and Hodgkin's disease (O/E = 7.69). Cancer risk, which did not vary according to initial treatment category, was also constant across all time intervals after CLL diagnosis. Conclusion: CLL patients are at a significantly increased risk of developing a second malignant neoplasm. The pattern of cancer excesses suggests a susceptibility state permitting the development of selected second malignancies in patients with CLL, perhaps because of shared etiologic factors, immunologic impairment, and/or other influences. Although our results do not suggest a strong treatment effect, more detailed studies of second tumors in CLL are needed to investigate the role of radiation therapy and chemotherapy. [J Natl Cancer Inst 84:1422-1427, 1992]
2
However, the number of patients in whom this sequence of diagnoses was reported is noted.
Vol. 84, No. 18, September 16, 1992
To calculate the risk of second cancers, we assembled person-years of observation according to age, sex, and calendar periods from 2 months after the date of CLL diagnosis to the date of last follow-up evaluation, date of death, date of diagnosis of a second cancer, or the end of the study (December 31, 1988), whichever occurred first. SEER incidence rates specific for 5-year age, sex, and 5-year calendar year intervals were multiplied by the accumulated person-years at risk to estimate the number of cancer cases expected. Statistical tests and 95% confidence limits were based on the assumption that the observed numbers of second tumors were distributed as a Poisson variable. The methods of Breslow et al. (29) were used to conduct tests for homogeneity and linear trend.
Results During the study period, 9456 twomonth survivors of CLL were reported to the SEER registries (Table 1). The average age at diagnosis was 70.0 years, and the average follow-up was 4.22 years. Approximately 92% of the patients were White, and 6.1% were Black. Second cancers developed in 8.9% of subjects. Diagnoses were histo-
logically or cytologically confirmed in 94.6% of the CLL cases and in 94.1% of the second tumors. Initial treatment is presented in Table 2. A second malignancy developed in 840 subjects compared with 658 expected (observed-to-expected ratio [O/E] = 1.28; 95% confidence interval [CI] = 1.19-1.37) (Table 3). Elevated risks were observed for cancers of the lung, eye, and brain and for malignant melanoma and Hodgkin's disease. All ocular tumors were melanomas. A significant excess of prostate cancer was limited to the 1st year after diagnosis of CLL, a pattern consistent with a surveillance effect. Nonsignificant elevations in cancers of the larynx, testis, kidney, and connective tissue were also apparent. The histology of the connective tissue tumors, all of which occurred in men, included malignant fibrous histiocytoma (n = 2), Kaposi's sarcoma (n = 1), and liposarcoma (n = 1). The median time for the development of connective tissue tumors following CLL was 33 months. Malignancies of the gastrointestinal tract were not significantly increased. Decreased risks for cancers of the breast, uterine cervix, and uterine corpus were apparent, although ovarian neoplasms occurred slightly in excess.
Downloaded from http://jnci.oxfordjournals.org/ at Brunel University on January 3, 2015
Detroit, Atlanta (1975-1988), SeattlePuget Sound (1974-1988), and San Francisco-Oakland as well as the states of Connecticut, Hawaii, Iowa, New Mexico, and Utah—areas which comprise approximately 10% of the U.S. population. Patient demographic data, vital status, and tumor histology are routinely reported to the SEER Program. In addition, SEER files also include the type of initial treatment given to patients according to one of several broad designations. For analyses, we grouped patients into the following categories: radiation therapy only, chemotherapy only, radiation therapy and chemotherapy, no treatment, and other treatment. The last category consisted of subjects whose initial therapy was either hormonal (including corticosteroids), immunologic, surgical, or unknown. The SEER Program does not record therapy given after the first course of treatment. In addition, information is not available with regard to the specific drugs administered or dose schedules used. However, patients with CLL are usually treated with continuous or pulse administration of low-dose chlorambucil with or without prednisone (21-23). Less frequently, either cyclophosphamide (24) or combination chemotherapy regimens (25) are used. Radiation therapy is generally used only to treat areas of bulky disease that do not respond to chemotherapy; splenic irradiation is given in selected patients (26,27). SEER files were examined for invasive primary malignancies that developed at least 2 months after the initial CLL diagnosis. Cancer sites were grouped according to the World Health Organization's International Classification of Diseases for Oncology (ICD-O) (28). Because reporting of second diagnoses of leukemia following CLL was incomplete during the early years of the SEER Program, such malignancies were not included in any evaluations. Also, since CLL may represent the leukemic phase of certain types of non-Hodgkin's lymphoma, the latter diagnosis was excluded from analyses of second cancer risk.2
Table 1. Characteristics of patients with CLL reported to the SEER cancer registries, 1973-1988 Category No. with CLL' No. who developed a second primary cancer Average age at diagnosis of CLL, y Person-years of follow-up Average follow-up, y
Men
Women
Total
5508 575 68.5 22 103 4.01
3948 265 72.0 17 840 4.51
9456 840 70.0 39943 4.22
* ICD-0 morphology code 9823. Number excludes all persons who survived
