Original article

Annals of Oncology 13: 318–322, 2002 DOI: 10.1093/annonc/mdf042

Is there an increased rate of additional malignancies in patients with mantle cell lymphoma? I. Barista, F. Cabanillas*, J. E. Romaguera, I. F. Khouri 1, Y. Yang2, T. L. Smith2, S. S. Strom3, L. J. Medeiros4 & F. B. Hagemeister Department of Lymphoma and Myeloma, 1Department of Blood and Marrow Transplantation, 2Department of Biostatistics, 3Department of Epidemiology and 4 Department of Pathology, University of Texas M.D. Anderson Cancer Center, Houston, TX, USA Received 31 August 2001; accepted 11 September 2001

Purpose: To examine the frequency of additional neoplasms preceding and following the diagnosis of mantle cell lymphoma (MCL).

Patients and methods: A total of 156 patients with MCL treated on the hyperfractionated cyclophosphamide, vincristine, doxorubicin and dexamethasone alternated with methotrexate and cytosine arabinoside (Hyper-CVAD/M-A) program with or without rituximab from 1994 to 2000 were the subjects of this report. Results: These patients were followed for a median time of 26 months, and a total of 32 (21%) additional neoplasms were diagnosed, 21 preceding the diagnosis of MCL and 11 following MCL. After excluding certain types of non-invasive neoplasms, including basal cell carcinoma, meningioma and cervical intraepithelial neoplasia, we observed seven second malignancies after the diagnosis of MCL, and the 5-year cumulative incidence rate of second malignancy was 11%. The observed-to-expected (O/E) ratio was 7/0.07 = 100 [95% confidence interval (CI) 49.3 to 186.6; P <0.0001]. Of the 21 malignancies diagnosed prior to MCL, 16 were invasive and five non-invasive. There were a total of 10 urologic malignancies occurring before or after the diagnosis of MCL was established. Conclusions: Our findings suggest that there is an increased incidence of second malignancies in patients with MCL. In addition, the high number of cases with urinary tract cancer in our series may substantiate prior reports describing a possible association between lymphoma and urologic malignancies. Key words: additional neoplasm, mantle cell lymphoma, second malignancy

Introduction Several studies, including some large cohorts, have reported an increased frequency of second malignancies in patients with lymphoma [1–3]. With the tremendous success in the management of Hodgkin’s disease (HD) over the past 20 years, the long-term risk of treatment began to emerge as a major hazard of that success. As a consequence, HD was the first model in which investigators studied the long-term carcinogenic effects of therapy. The risk of a second cancer following therapy for non-Hodgkin’s lymphoma (NHL) has been studied less extensively than that following HD, in part due to the older age and less favorable prognosis of patients with NHL. However, patients with HD or NHL also may be at

*Correspondence to: Department of Lymphoma and Myeloma, University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030-4009, USA. Tel: +1-713-792-2860; Fax: +1-713-794-5656; E-mail: [email protected] © 2002 European Society for Medical Oncology

inherent risk for second malignancies, and not all of these cancers are necessarily due to therapy. Mantle cell lymphoma (MCL), although described by others over 20 years ago using different terms, was re-named MCL by consensus in 1992 [4], the name being retained in the Revised European American Lymphoma (REAL) and World Health Organization (WHO) classifications. The median survival for patients with MCL ranges from 2 to 5 years with standard cyclophosphamide, doxorubicin, vincristine, prednisone (CHOP) therapy. Intensive therapies, transplantation, and monoclonal antibodies are new avenues of research in its management. At our institution, we have observed improved survival for patients with MCL treated with the hyperfractionated cyclophosphamide, vincristine, doxorubicin and dexamethasone alternated with methotrexate and cytosine arabinoside (Hyper-CVAD/M-A) regimen. It has been our impression that many patients with MCL have developed other neoplasms, aside from MCL. This prompted us to investigate the issue of second neoplasms in a group of 156 patients with MCL treated with two protocols at

319 our institution. To our knowledge, no previous study has specifically addressed the issue of second malignant neoplasms in patients with MCL. In this study, we examined the frequency of additional neoplasms preceding and following the diagnosis of MCL. Our secondary objective was to determine whether the number of second malignancies observed in our patient population was greater than the expected number, based on the incidence of all cancers calculated from the Connecticut Tumor Registry incidence data.

Patients and methods From 1994 to 2000, 156 adult patients with MCL were accrued on two consecutive protocols (DM 94-130 and DM 98-327) at the University of Texas M.D. Anderson Cancer Center (Houston, Texas). The median age at time of diagnosis was 59.5 years, with a range from 37 to 81 years. Of the patients in the study 119 were male (76%) and 37 were female (24%). All patients had a pathologically confirmed diagnosis of MCL at our center using strict morphological criteria. The biopsy specimens were assessed by immunophenotypic methods, including antibodies specific for CD5, CD20 and cyclin D1, and when possible with either cytogenetics, polymerase chain reaction (PCR) to amplify the t(11;14) involving the major translocation cluster region of the bcl-1 locus, or fluorescence in situ hybridization (FISH) studies for chromosome 11q breakpoints and/ or chromosome 11 and 14 probes for the t(11;14). The following information on each patient was retrospectively obtained from their medical records: history, physical examination, complete blood count, serum chemistry panel, chest radiograph, computerized tomography of abdomen and pelvis, endoscopies of upper and lower gastrointestinal tracts [5], and bilateral bone marrow aspirations and biopsies. Stage of disease was determined using the Ann Arbor staging system. The first treatment trial [6] consisted of Hyper-CVAD/M-A chemotherapy. Young patients who achieved a complete or good partial response after four cycles of chemotherapy then received total body irradiation (TBI), high-dose cyclophosphamide and stem cell transplantation (SCT). Older patients received eight cycles of Hyper-CVAD/M-A. In the second trial [7], we added the anti-CD20 antibody, rituximab, 24 h before each of the first six cycles of chemotherapy. In the latter study, patients who entered complete remission after two cycles of Hyper-CVAD/M-A plus rituximab received six cycles of chemotherapy without consolidation with TBI and high-dose therapy (HDT) with SCT. Patients who were still in partial remission after six cycles underwent HDT with SCT. We identified those cases with additional neoplasms by review of the medical records of each patient, and considered only patients with histologically proven additional neoplasms. In accordance with most studies, we excluded non-melanoma skin cancers, in situ cancers and other noninvasive tumors such as meningiomas. The time at risk was defined as the time from diagnosis of MCL to occurrence of a second malignancy, death or last contact. The main source of information was the hospital chart, but when necessary, additional follow-up information was obtained from the patient or the patient’s family.

secondary invasive cancers was determined using sex, age and calendaryear specific rates from the Connecticut Tumor Registry for the period from 1936 to 1985 applied to the relative person-years at risk. The relative risk of developing a second cancer was estimated by the ratio of the observed (O) number of second cancer cases to the number expected (E). Tests of significance for the standardized cancer incidence ratios (O/E) were calculated assuming the observed number of second malignancies followed a Poisson distribution. The computer program Cohort Analysis for Genetic Epidemiology (CAGE) was used for these calculations [8] . We plotted the cumulative incidence of any second malignancy, using a method for calculation of cause-specific cumulative incidence [9] . The reverse Kaplan–Meier method [10] was not preferred, since it does not properly account for the reduced probability due to competing risks of death attributable to MCL, and thus, may overestimate the incidence of second malignancy.

Results Of the 119 (76%) patients in this study who were male, 141 (90%) had stage IV disease (Table 1). Histologically, 117 (75%) patients had nodular and/or diffuse neoplasms. Five (3%) had a mantle zone pattern. All patients received an alkylating agent and an anthracycline-containing regimen. Furthermore, 16 patients (10%) received purine analogs at sometime during their therapy. Fifty patients (32%) received TBI as a part of their preparative regimen. After HDT, 40 (26%), 16 (10%) and two (1%) patients received autologous, allogeneic, and both treatments at different times, respectively. The median follow-up was 26 months. At 5 years, 98 (63%) patients were projected alive, and 40 (25%) were projected to

Table 1. Characteristics of 156 patients with mantle cell lymphoma Age

Median (range)

Gender

Male

119 (76%)

Race

Caucasian

142 (91%)

Pathology

Mantle zone Nodular Diffuse ± Nodular

Stage

Protocol

Therapy

59.5 (37–81)

5 (3%) 12 (8%) 105 (67%)

Blastic

19 (12%)

Not specified

15 (10%)

I & II & III

15 (10%)

IV

141 (90%)

94-130

100 (64%)

98-327

56 (36%)

Alkylating agent

156 (100%)

Anthracycline

156 (100%)

Statistical considerations

Purine analogs

16 (10%)

To determine whether there was an excess of secondary cancers after MCL diagnosis, we computed standardized incidence ratios calculated by determining the ratio of the observed to expected number of individuals with second malignancies. The expected number of individuals with

Transplant

58 (37%)

Radiotherapy

62 (40%)a

a

Radiotherapy: total body irradiation, 50; conventional, 12.

320

Table 2. Second malignant neoplasms occurring before and after the diagnosis of mantle cell lymphomaa

a

Months

Neoplasm

–312

Thyroid

–240

Melanoma

–154

Breast

–144

Renal

–132

Low-grade non-Hodgkin’s lymphoma

–96

Prostate

–90

Colon

–80

Bladder

Figure 1. Cumulative incidence of second malignancies (non-invasive

–34

Prostate

neoplasms are excluded).

–33

Melanoma

–30

Prostate

–27

Gastric

–26

Thyroid

–18

Renal

–3

Prostate

–1

Chronic myeloid leukemia

0

Diagnosis of mantle cell lymphoma

1

Prostate

7

Renal

10

T-cell non-Hodgkin’s lymphoma

37

Acute myelomonocytic leukemia

40

Lung

57

Renal

89

Myelodysplastic syndrome (monosomy 7)

Sorted according to time in months.

be disease-free. Twenty-seven (17%) had died: 24 died due to progression of lymphoma or infections, and three died as a consequence of a second malignancy. A total of 32 (21%) additional neoplasms were diagnosed, 21 preceding the diagnosis of MCL and 11 following MCL. Excluding the non-invasive additional neoplasms, 16 patients (10%) had malignancies prior to the diagnosis of MCL, whereas seven (4.5%) had a second malignancy following the diagnosis of MCL. Most of the latter occurred soon after the diagnosis of MCL; however, long latencies occurred between malignancies in the former group. Sites of second malignancies are listed in Table 2 along with the interval between diagnosis of MCL and the detection of the second malignancy. After omitting non-invasive tumors, the cumulative incidence of second malignancy at 5 years following the diagnosis of MCL was 11% (Figure 1). Regarding those second malignancies occurring after the diagnosis of MCL, there was a 100-fold excess of secondary malignancies. We observed seven cases, the number of

Figure 2. Cumulative incidence of second malignancies per Travis article [3].

expected second malignancies was 0.07, and the standardized cancer incidence ratio was 100 [95% confidence interval (CI) 49.3 to 186.6; P <0.0001]. With regard to the more specific question about incidence rates in the general lymphoma population, we attempted to apply the methods of Travis et al. [3] to our data set. We excluded those second malignancies occurring within the first 2 years after diagnosis, as well as ‘non-invasive’ cancers, and calculated the incidence on a ‘reverse’ Kaplan–Meier curve (Figure 2). This resulted in a total of four second malignancies. The 5-year incidence rate was 11% (95% CI 0% to 24%) whereas this rate was 3% in the Travis study.

Discussion We observed an unexpected number of second malignancies, particularly genitourinary tumors, both prior to and after the diagnosis of MCL. Reasons for this observation are unclear, but it suggests either a genetic predisposition or some other common cause for both tumor groups. The age of the patients possibly may be another contributing factor, since MCL is uncommon before the age of 50 [11]. Although the standardized cancer incidence ratio of the second malignancies occurring

321 after the diagnosis of MCL was considerably higher than expected, and the observation was statistically significant, it would be desirable to confirm this in a larger cohort of cases. To our knowledge, there are no studies specifically addressing the issue of second malignancies in MCL. We were able to find two studies on MCL which mention second malignancy. In the first study of 121 patients, the investigators observed 10 malignancies prior to and four after MCL [11]. Regarding the neoplasms that followed MCL, the median time for onset of second malignancy was 20 months (range 12–84). In our study, this period was 37 months (range 1–89). The second study evaluated outcome of 24 MCL patients receiving autologous transplants of whom 13 also received TBI. Five patients developed secondary malignancies (two acute myeloid leukemia and three solid tumors) 8, 15, 30, 30 and 40 months after transplant [12]. They emphasized that the incidence of secondary malignancies was higher in this group than generally reported, and that HDT and/or radiotherapy had carcinogenic potential. Travis et al. [2, 3, 13] have published a series of reports which included thousands (29 153 and 6171) of patients diagnosed with NHL as the primary cancer who survived long enough to be evaluable for the incidence of second malignancies occurring after this disorder. These authors reported that the increased risks persisted for all second cancers (including renal cell and bladder carcinomas) among 15-year survivors of NHL (O/E = 1.45). The actuarial risk of developing a second cancer 3–20 years after the diagnosis of NHL was 21.1%, compared with a population-expected cumulative risk of 15.4% [3]. Our results are in accordance with the 5-year results of that study. Among the long-term (15-year) survivors in the study of Travis et al. [3], elevated risks of second cancers were restricted to men (O/E = 1.79). In the case of MCL, it is difficult to corroborate this observation, since this disorder predominantly affects males; 76% of patients in the current study were men. There were three patients in our series with second malignancies occurring within the first year after diagnosis of MCL. Various mechanisms have been postulated for the occurrence of synchronous and metachronous second neoplasms in patients with lymphoma, unrelated to therapy. First, the detection of a second neoplasm may be a consequence of early detection by procedures used in the diagnosis, staging and subsequent evaluation of response to treatment for the first malignancy. Moreover, elevated risks of dual malignancy may reflect the effects of host susceptibility or shared etiological factors. In our series, the numbers were insufficient to make a formal comparison of the second malignancies according to TBI status. In general, SCT recipients are at increased risk of developing a later malignancy [14]. The risk of myelodysplasia (MDS) as a complication of cytotoxic chemotherapy is particularly associated with alkylating agents and etoposide [15, 16]. Repeated and prolonged episodes of drug-induced

bone marrow dysplasia and genetic alterations in bone marrow stem cells are considered to be the main possible causes. In the case of lymphoma, considerably higher risks of acute leukemia were found in studies which evaluated treatment strategies consisting of long-term chemotherapy administration and TBI [17–21]. However, HDT regimens followed by SCT may increase the risk of a second malignancy regardless of whether TBI is utilized [22]. We observed a total of 10 urologic malignancies (five prostate, four renal cell and one bladder carcinoma), seven before and three after the diagnosis of MCL. A report by Nishikubo et al. [23] has previously suggested a relationship between renal cell carcinoma (RCC) and lymphoid malignancies. In their study population of 186 patients with RCC and 405 patients with lymphoid malignancies, there were eight cases with both conditions; in four the diagnosis of RCC preceded that of the hematological malignancy. Likewise, we had four cases of RCC and MCL, two occurring prior to and two after MCL diagnosis. One of our patients was diagnosed as having RCC 7 months after the diagnosis of MCL, making the possibility of a treatment-related malignancy extremely unlikely. Furthermore, in a previous study from our institution, we have also reported a higher than expected incidence (relative risk 2.7) of RCC with NHL [24]. In that study, there were 41 patients who had both conditions simultaneously or at different times. We had only one case of bladder carcinoma and it occurred 80 months prior to the diagnosis of MCL. Travis and associates have identified a significant 4.5-fold higher risk of bladder cancer following therapy with cyclophosphamide in 2-year survivors of NHL, and the risk was dependent upon cumulative dose [13]. Radiotherapy given without cyclophosphamide was not associated with a significantly increased risk of bladder malignancy. In a recent study from the same group, 43% of cyclophosphamide-associated bladder tumors also had p53 mutations [25]. The relatively short follow-up, and the lack of a welldefined control group are the major shortcomings of the present study. Although the power of the present study is insufficient to show an increased incidence of additional tumors in MCL compared with other NHLs, there appears to be a statistically significant increase in the number of additional neoplasms occurring in patients with MCL compared with the general population. Continued follow-up of these patients is important to determine whether the risk continues to increase with time. Furthermore, the occurrence of 10 urinary tract malignancies in our study suggests a possible association between these disorders and lymphomas [23, 24, 26].

References 1. Berg JW. The incidence of multiple primary cancers. I. Development of further cancers in patients with lymphomas, leukemias, and myeloma. J Natl Cancer Inst 1967; 38: 741–752. 2. Travis LB, Curtis RE, Boice JD et al. Second cancers following nonHodgkin’s lymphoma. Cancer 1991; 67: 2002–2009.

322 3. Travis LB, Curtis RE, Glimelius B et al. Second cancers among longterm survivors of non-Hodgkin’s lymphoma. J Natl Cancer Inst 1993; 85: 1932–1937. 4. Banks PM, Chan J, Cleary ML et al. Mantle cell lymphoma. A proposal for unification of morphologic, immunologic, and molecular data. Am J Surg Pathol 1992; 16: 637–640. 5. Romaguera J, Dang N, Hagemeister FB et al. Evidence for need of upper and lower endoscopies with biopsy in staging of asymptomatic patients with aggressive mantle cell lymphoma (MCL). Blood 2000; 96: 334a (Abstr 1440). 6. Khouri IF, Romaguera J, Kantarjian H et al. Hyper-CVAD and highdose methotrexate/cytarabine followed by stem-cell transplantation: an active regimen for aggressive mantle cell lymphoma. J Clin Oncol 1998; 16: 3803–3809. 7. Romaguera JE, Dang N, Hagemeister FB et al. Preliminary report of Rituximab with intensive chemotherapy for untreated aggressive mantle cell lymphoma (MCL). Blood 2000; 96: 733a (Abstr 3170). 8. Lustbader ED, McLaughin PR. CAGE: Cohort Analysis for Genetic Epidemiology. Philadelphia: Fox Chase Cancer Center 1995. 9. Kalbfleisch JP, Prentice RL. The statistical analysis of failure time data. New York: Wiley 1980. 10. Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481. 11. Samaha H, Dumontet C, Ketterer N et al. Mantle cell lymphoma: a retrospective study of 121 cases. Leukemia 1998; 12: 1281–1287. 12. Decaudin D, Brousse N, Brice P et al. Efficacy of autologous stem cell transplantation in mantle cell lymphoma: a 3-year follow-up study. Bone Marrow Transplant 2000; 25: 251–256. 13. Travis LB, Curtis RE, Glimelius B et al. Bladder and kidney cancer following cyclophosphamide therapy for non-Hodgkin’s lymphoma. J Natl Cancer Inst 1995; 87: 524–530. 14. Bhatia S, Ramsay NK, Steinbuch M et al. Malignant neoplasms following bone marrow transplantation. Blood 1996; 87: 3633–3639. 15. Kantarjian HM, Keating MJ, Walters RS et al. Therapy-related leukemia and myelodysplastic syndrome: clinical, cytogenetic, and prognostic features. J Clin Oncol 1986; 4: 1748–1757. 16. Karp JE, Smith MA. The molecular pathogenesis of treatmentinduced (secondary) leukemias: foundations for treatment and prevention. Semin Oncol 1997; 24: 103–113.

17. Greene MH, Young RC, Merrill JM, DeVita VT. Evidence of a treatment dose response in acute nonlymphocytic leukemias which occur after therapy of non-Hodgkin’s lymphoma. Cancer Res 1983; 43: 1891–1898. 18. Gomez GA, Aggarwal KK, Han T. Post-therapeutic acute malignant myeloproliferative syndrome and acute nonlymphocytic leukemia in non-Hodgkin’s lymphoma. Cancer 1982; 50: 2285–2288. 19. Travis LB, Weeks J, Curtis RE et al. Leukemia following low-dose total body irradiation and chemotherapy for non-Hodgkin’s lymphoma. J Clin Oncol 1996; 14: 565–571. 20. Darrington DL, Vose JM, Anderson JR et al. Incidence and characterization of secondary myelodysplastic syndrome and acute myelogenous leukemia following high-dose chemoradiotherapy and autologous stem-cell transplantation for lymphoid malignancies. J Clin Oncol 1994; 12: 2527–2534. 21. Milligan DW, Ruiz De Elvira MC, Kolb HJ et al. Secondary leukaemia and myelodysplasia after autografting for lymphoma: results from the EBMT. EBMT Lymphoma and Late Effects Working Parties. European Group for Blood and Marrow Transplantation. Br J Haematol 1999; 106: 1020–1026. 22. Oddou S, Vey N, Viens P et al. Second neoplasms following highdose chemotherapy and autologous stem cell transplantation for malignant lymphomas: a report of six cases in a cohort of 171 patients from a single institution. Leuk Lymphoma 1998; 31: 187–194. 23. Nishikubo CY, Kunkel LA, Figlin R et al. An association between renal cell carcinoma and lymphoid malignancies. A case series of eight patients. Cancer 1996; 78: 2421–2426. 24. Anderson CM, Pusztai L, Palmer JL et al. Coincident renal cell carcinoma and non-Hodgkin’s lymphoma: the M.D. Anderson experience and review of the literature. J Urol 1998; 159: 714–717. 25. Khan MA, Travis LB, Lynch CF et al. p53 mutations in cyclophosphamide-associated bladder cancer. Cancer Epidemiol Biomarkers Prev 1998; 7: 397–403. 26. Tihan T, Filippa DA. Coexistence of renal cell carcinoma and malignant lymphoma. A causal relationship or coincidental occurrence? Cancer 1996; 77: 2325–2331.