- Email: [email protected]
Characterizing Autoimmune Disease-associated Diffuse Large B-cell Lymphoma in a SEER–Medicare Cohort
Original Study
Characterizing Autoimmune Disease-associated Diffuse Large B-cell Lymphoma in a SEEReMedicare Cohort Jean L. Koff,1 Ashish Rai,2 Christopher R. Flowers1 Abstract We used the Surveillance, Epidemiology, and End Results (SEER)eMedicare-linked database to characterize the patterns of presentation, treatment, and survival in older diffuse large B-cell lymphoma (DLBCL) patients with concomitant autoimmune disease. DLBCL patients with autoimmune disease were more likely to be female but otherwise did not differ significantly from other DLBCL patients in demographic data, treatment, or clinical outcomes. Background: Severe immune dysregulation such as seen in autoimmune (AI) disease is known to act as a significant risk factor for diffuse large B-cell lymphoma (DLBCL). However, little is known about the demographics or clinical outcomes of DLBCL that arises in the setting of AI disease. Patients and Methods: We used the Surveillance, Epidemiology, and End Results (SEER) database for patients with a diagnosis from 1999 to 2009 linked to their Medicare claims data through 2011 to characterize the presentation, treatment, and survival patterns in DLBCL patients, including those with rheumatoid arthritis, systemic lupus erythematosus (SLE), Sjögren syndrome, and other B-cellemediated AI diseases. We examined the baseline clinical characteristics for patients with B-celle mediated AI disease, plotted the overall survival and lymphoma-related survival (LRS) for these groups, and compared the median survival times. Results: Patients with DLBCL and AI disease were more commonly female. However, patients with DLBCL and rheumatoid arthritis, SLE, Sjögren syndrome, or other B-cell AI diseases did not differ from other DLBCL patients in any other baseline presenting features and received similar first-line treatment. A trend toward decreased LRS was seen in patients with SLE and DLBCL compared with all other groups, but this difference was not statistically significant in this cohort. Conclusion: In the present retrospective claims-based cohort of older patients with DLBCL, concomitant AI disease was uncommon and was more likely to occur in female DLBCL patients, which likely reflects the greater incidence of AI disease in women. The possibility of lower LRS for SLE patients should be explored in future studies. Clinical Lymphoma, Myeloma & Leukemia, Vol. -, No. -, --- ª 2017 Elsevier Inc. All rights reserved. Keywords: DLBCL, Non-Hodgkin lymphoma, Rheumatoid arthritis, Sjögren syndrome, Systemic lupus erythematosus
Introduction Non-Hodgkin lymphoma (NHL) is the most common hematologic malignancy worldwide.1 Its most frequently diagnosed subtype, diffuse large B-cell lymphoma (DLBCL), accounts for w25% to 30% NHL cases in Western countries.2,3 Importantly, although 1 Department of Hematology and Medical Oncology, Winship Cancer Institute of Emory University, Atlanta, GA 2 American Cancer Society, Atlanta, GA
Submitted: Jun 20, 2017; Revised: Nov 17, 2017; Accepted: Nov 27, 2017 Address for correspondence: Jean L. Koff, MD, Department of Hematology and Medical Oncology, Winship Cancer Institute of Emory University, 1365 Clifton Road, Building B, Suite 4300, Atlanta, GA 30322 Fax: (404) 778-3366; e-mail contact: [email protected]
2152-2650/$ - see frontmatter ª 2017 Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.clml.2017.11.009
DLBCLs appear histologically similar, they demonstrate a wide degree of genetic and clinical heterogeneity. DLBCL can be curable; however, nearly 50% of patients will eventually relapse, with subsequent dismal prognosis.4,5 Thus, substantial research efforts have been focused on identifying the biologic and clinical features that underlie these stark differences in outcomes with the hope that targeted therapeutic approaches could abrogate such disparities. Severe immune dysregulation such as that seen with human immunodeficiency virus infection, immunosuppression after solid organ transplantation, and autoimmune (AI) disease is known to act as a major risk factor for NHL. The InterLymph Subtypes Project pooled cases and controls to provide well-powered comparisons of risk factors for specific NHL subtypes, including DLBCL. Multivariate analysis of 4667 DLBCL cases and 22,639
Clinical Lymphoma, Myeloma & Leukemia Month 2017
-1
Mortality and Vascular Events Among Elderly Patients With CML controls showed that B-celleactivating AI diseases in general (odds ratio, 2.4; 95% confidence interval, 1.8-3.1), and systemic lupus erythematosus (SLE) and Sjögren syndrome (SS) in particular, were the most strongly associated with increased DLBCL risk after controlling for all other risk factors.6 These findings mirror those of large cohort studies of patients with AI disease. SS has long been known to carry the greatest risk of NHL (relative risk, 4-40 times that of the general population), most of which are mucosaassociated lymphoid tissue lymphomas.7 However, DLBCL cases comprised about 15% of SS-associated lymphoma in a 584-patient cohort.8 Similarly, a 2005 cohort study involving 9547 SLE patients found that aggressive subtypes predominated among the NHL diagnoses, with DLBCL constituting more than one half of the cases for which subtype was specified.9 The significant heterogeneity in the studies of NHL risk in patients with rheumatoid arthritis (RA) preempted the InterLymph group from performing focused analyses of NHL subtype risk in that population.10 However, a large Swedish study that included > 74,000 RA patients found that patients with greater indexes of inflammatory activity exhibited an increased risk of NHL in general and DLBCL in particular (48% of cases).11 Given that the processes of inflammation and chronic selfantigen stimulation that define AI diseases represent specific pathways that could promote lymphoma development, it is possible that AI-associated lymphomas are a distinct subset within DLBCL exhibiting characteristic clinical and biologic behavior. However, little is known about the demographic data or clinical outcomes of DLBCL that arises in the setting of AI disease. We examined the Surveillance, Epidemiology, and End Results (SEER)eMedicarelinked database to determine the frequency of common B-cell AI diseases among older patients with DLBCL and to characterize the patterns of presentation, treatment, and survival for DLBCL patients with concomitant AI disease.
Patients and Methods Data Source
2
-
We used the National Cancer Institute (NCI) SEER database for patients with a diagnosis from 1999 to 2009 linked to their Medicare claims data through 2011 to characterize the presentation, treatment, and survival patterns in patients with DLBCL, including those with RA, SLE, SS, and other B-cellemediated AI diseases as defined by the InterLymph criteria (AI hemolytic anemia, Hashimoto’s thyroiditis/hypothyroidism, myasthenia gravis, and pernicious anemia12). The SEER program reports data on cancer incidence and survival collected from US registries, covering about 28% of the population as of 2016.13 The collected data include patient demographics, tumor pathology, disease stage, primary tumor site, first-line treatment, and dates of diagnosis and death. Linking SEER with the Medicare claims data allows for the identification of concomitant health conditions and specific treatments received by elderly cancer patients. Among individuals > 65 years, 97% are Medicare-eligible, and 93% of those listed in SEER are linked to the Medicare enrollment file.14 Because this database does not include patient identifiers, our study did not require approval from an institutional review board. However, a data use agreement was signed before initiation of the study.
Clinical Lymphoma, Myeloma & Leukemia Month 2017
Eligibility Criteria Patients were considered eligible for analysis if they had received a diagnosis of DLBCL from January 1, 2002 to December 31, 2009, had linked Medicaid claims data available 2011, and were aged 66 years at diagnosis. The minimum required age was 66 years to ensure that patients had been enrolled in Medicare for 12 months before the diagnosis. Cases were identified using the World Health Organization International Classification of Diseases for Oncology, 3rd edition, histology codes 9680 and 9684.15 The following International Classification of Diseases, 9th revision, Clinical Modification codes were used to identify the presence of concomitant AI disease: SLE, 710.0; RA, 714.0 to 714.3; AI hemolytic anemia, 283.0; myasthenia gravis, 358.00 and 358.01; pernicious anemia, 281.0; and SS: 710.2. The exclusion criteria are shown in Figure 1.
Patient Characteristics Patients were stratified into groups by AI disease (none coded, SLE, SS, RA, or other B-celleactivating AI disease) as identified in Medicare claims. Self-reported race was categorized as white, African-American, or other; using SEER data, the “other” category includes individuals of Asian, Native American, Pacific Islander, or Alaskan Native ancestry.16 The SEEReMedicare database uses census tract information (eg, percentage of residents living in poverty and percentage with only a high school education) from the 2000 US Census as a surrogate for socioeconomic status, as described in other SEEReMedicare studies.17-19 Other demographic variables analyzed in the present study included sex, marital status, and type of geographic area (less urban or rural, urban, or metropolitan). In terms of disease status, patients were classified with regard to the following: Ann Arbor stage (I/II, III/IV, or unknown), primary disease site (nodal vs. extranodal), presence of B symptoms, performance status, NCI comorbidity index score (0, 1, or 2), and year of diagnosis. Performance status was classified as poor if a patient had claims for any of the following: hospice, home health agency, skilled nursing facility, oxygen, or wheelchair or related supplies. Such claims-based measures of performance status have been used in other cancer studies.20-23 The NCI comorbidity index scores were calculated using the Deyo adaptation of the Charlson comorbidity index to identify the 15 noncancer comorbidities included in the Charlson comorbidity index from the Medicare claims.24,25
Treatment and Mortality Classification We determined the initial management strategies using the Medicare claims made within 6 months of diagnosis; if no treatment was documented within that period, management was categorized as “observation.” The SEEReMedicare data set does not include information regarding the receipt of oral medications without an intravenous equivalent. Thus, patients with claims for cyclophosphamide, doxorubicin, and vincristine were categorized as having received CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone), and those with claims for cyclophosphamide and vincristine were classified as having received CVP (cyclophosphamide, vincristine, prednisolone). Patients with those same claims who also received rituximab were classified as having received R-CHOP and R-CVP, respectively.
Jean L. Koff et al Figure 1 Selection of the Study Cohort
All DLBCL patients in the SEER-Medicare database diagnosed from 1999 through 2009 n=33,886
Less patients with unknown age at diagnosis or diagnosis after death or SEER and Medicare reporting different dates of death n=469
DLBCL patients with known age at diagnosis n=33,417
DLBLC patients in patients aged 66 and older n=24,387
Less patients with age at diagnosis less than 66 years n=9,030 Less patients with interrupted Medicare coverage and/or HMO enrolment 12 months before and 6 months after diagnosis n=12,930
DLBCL patients with adequate Medicare coverage n=11,457
Less patients that were not histologically confirmed n=512
Adequately diagnosed DLBCL patients n=10,945
Less patients with date of initiation of therapy more than 45 days before date of diagnosis n=506
Eligible DLBCL patients initially observed or treated n=10,439 Less patients initially treated with regimens other than R-CHOP, R-CVP, CHOP, or CVP n=4,515 Study cohort of DLBCL patients n=5,924 DLBCL patients with rheumatoid arthritis n=155 DLBCL patients with systemic lupus erythematosus n=25
DLBCL patients with any B-cell autoimmune disease n=270
DLBCL patients with Sjogren's syndrome n=18
Abbreviations: CHOP ¼ cyclophosphamide, doxorubicin, vincristine, prednisone; CVP ¼ cyclophosphamide, vincristine, prednisolone; DLBCL ¼ diffuse large B-cell lymphoma; HMO ¼ health maintenance organization; R ¼ rituximab; SEER ¼ Surveillance, Epidemiology, and End Results.
We used the SEER date and cause of death to determine the mortality classification. Because SEER only reports diagnoses by month and year, the date of diagnosis for our survival analyses was designated as the 15th day of the reported month. Patients were followed up until death, enrollment in a health maintenance organization, or last date of available Medicare claims. We examined 2 different survival endpoints: overall survival (OS), measured from the date of diagnosis until death, censored at the last follow-up; and lymphoma-related survival (LRS), measured from the date of diagnosis until death from lymphoma, censored at death from other causes or the last follow-up.
Statistical Analysis Patients without concomitant AI disease were compared to patients with SLE, SS, RA, or any B-cellemediated AI disease using c2 tests. Multivariable logistic regression models were used to investigate the relationships between patient characteristics and treatment regimen, and odds ratios with 95% confidence intervals (CIs) were calculated. Multivariable regression models were adjusted for the following demographic and clinical variables: sex, race, marital status, percentage in the census tract living in poverty, percentage in the census tract with only a high school education, type of geographic area, disease stage, primary disease site, presence of B symptoms, NCI comorbidity index score, performance status,
and year of diagnosis. We used the Hosmer-Lemeshow test to assess the goodness of fit of the logistic regression model, which was found to fit the data well. Kaplan-Meier curves were constructed to examine the effect of concurrent AI disease diagnosis on OS and LRS. Cox proportional hazards models were adjusted for the same variables described. We tested the global proportional hazards assumption with the Wald test, and proportional hazards assumptions for individual covariates were tested by assessing Schoenfeld residuals. No violations were detected. Sensitivity analyses were performed using propensity score methods to adjust for imbalances in observable covariates between treatment groups. We set a at 0.05 to determine the statistical significance, and all P values were 2-sided. Data were analyzed using SAS, version 9.4 (SAS Institute, Cary, NC), and Stata, version 13 (StataCorp LP, College Station, TX).
Results We identified 5926 patients with DLBCL, of whom 270 had B-cellemediated AI disease. The patient characteristics are summarized in Table 1. The SEEReMedicare data use agreement stipulates that patient data with numbers < 11 should not be directly reported or be derivable with more precision than “n < 11,” which limits the presentation of baseline characteristics for patients with concurrent DLBCL and either RA (n ¼ 155), SLE (n ¼ 25),
Clinical Lymphoma, Myeloma & Leukemia Month 2017
-3
Mortality and Vascular Events Among Elderly Patients With CML Table 1 Demographic and Disease Characteristics of DLBCL Patients Stratified by AI Disease in SEEReMedicare Database, 1999-2009 Characteristic Total patients
All Patients
B-cell AI Disease
5924
270
Age, y Median
75
75
71-80
71-79
Male
2811 (47)
79 (29)
Female
3113 (53)
191 (71)
IQR Sex
Table 2 Cox Regression Analysis of Overall Survival Stratified by Stage AI Disease
All Stages
Stage I/II
Stage III/IV
None RA SLE SS Other B-cell AI disease
Reference 0.86 (0.69-1.08) 1.17 (0.70-1.95) 1.23 (0.68-2.23) 1.22 (0.90-1.64)
Reference 0.82 (0.60-1.12) 0.97 (0.40-2.37) 1.07 (0.44-2.63) 1.21 (0.82-1.79)
Reference 0.83 (0.58-1.19) 1.22 (0.60-2.48) 1.59 (0.65-3.87) 1.25 (0.77-2.04)
Data presented as hazard ratio (95% confidence interval). Abbreviations: AI ¼ autoimmune; RA ¼ rheumatoid arthritis; SLE ¼ systemic lupus erythematosus; SS ¼ Sjögren syndrome.
Race White
5341 (90)
243 (90)
African American
185 (3)
NA
Other
398 (7)
NA
Stage I/II
3224 (54)
147 (54)
III/IV
2290 (39)
107 (40)
410 (7)
16 (6)
Nodal
3820 (64)
168 (62)
Extranodal
2104 (36)
102 (38)
Absent
4835 (82)
211 (78)
Present
1089 (18)
59 (22)
0
3738 (63)
48 (18)
1
1422 (24)
119 (44)
764 (13)
103 (38)
Unknown Nodal status
Poor performance status
NCI comorbidity index
>2 Frontline management CHOP
1025 (17)
47 (17)
R-CHOP
3346 (57)
157 (58)
122 (2)
NA
CVP R-CVP No recorded chemotherapy
345 (6)
NA
1086 (18)
42 (16)
OS for R-CHOP and CHOP patients Median
8.0
7.1
95% CI
7.7-8.4
5.0-8.7
R-CHOP, R-CVP, CHOP, or CVP), 40% had missing data for the agents used, 7% contained both rituximab and an anthracycline, 24% included rituximab but no anthracycline, 3% included an anthracycline without rituximab, and 7% contained neither rituximab nor an anthracycline; 19% of the patients had received radiation alone. The total number of chemotherapy cycles received did not vary significantly by AI disease status, with 54% of all patients receiving 4 cycles compared with 58% of RA, 68% of SLE, and 60% of SS patients. Similarly, 31% of all patients received 6 cycles compared with 38% of RA, 45% of SLE, and 33% of SS patients. The median OS for patients with AI disease did not differ significantly from that seen for patients without AI disease. The median OS was 6.34 years (95% CI, 4.55-8.60 years) for RA patients, 3.23 years (95% CI, 2.14-11.97 years) for SLE patients, 5.51 years (95% CI, 2.42 years to not reached) for SS patients, and 5.97 years (95% CI, 4.45-7.78 years) for patients with any Bcellemediated AI disease compared with 6.96 years (95% CI, 6.697.37 years) for patients without AI disease. A trend toward decreased LRS was seen in patients with SLE and DLBCL compared with other groups; however, this difference was not statistically significant in this cohort (Table 2 and Figure 2). The median LRS was not reached for any group. Because few differences were found in the baseline characteristics among the groups, traditional multivariable regression models are reported rather than propensity-matched models.
Discussion Data presented as n (%). Abbreviations: AI ¼ autoimmune; CHOP ¼ cyclophosphamide, doxorubicin, vincristine, prednisone; CI ¼ confidence interval; CVP ¼ cyclophosphamide, vincristine, prednisone; IQR ¼ interquartile range; NA ¼ not available (data for which n < 11 limited direct reporting per the SEEReMedicare data use agreement); NCI ¼ National Cancer Institute; OS ¼ overall survival; R ¼ rituximab.
4
-
or SS (n ¼ 18). With the exception that patients with concomitant DLBCL and AI disease were more commonly female, patients with DLBCL and RA, SLE, SS, or other B-cell AI diseases had baseline presenting features similar to those of other DLBCL patients and received similar first-line treatment. More specifically, no statistically significant differences were found in the receipt of either anthracycline-containing regimens (74% of all patients, 78% of RA, 80% of SLE, 83% of SS) or rituximab (63% of all patients, 66% of RA, 60% of SLE, 72% of SS). Among the treatment regimens excluded from analysis (ie, regimens other than chemotherapy with
Clinical Lymphoma, Myeloma & Leukemia Month 2017
Our results serve as an initial characterization of AI-associated DLBCL in older patients. By virtue of the use of Medicare claims to capture data on AI diagnosis and cancer treatments, our analysis was limited to patients > 65 years. Thus, we would expect to capture more cases of de novo and AI-associated DLBCL by analysis of data sets that include younger patients. Our group has previously shown that African-American DLBCL patients present at a younger age and with more aggressive disease compared with white patients.26 Reported data have suggested that such a relationship might exist for patients with AI disease as well. For example, the median age of SLE patients at NHL diagnosis was 49 years in 1 study of 11 cases27 and 57 years in a 2005 study that included 42 cases.9 Also, a 2012 study of SS-associated NHL reported a median age at DLBCL diagnosis of 66.5 years.8 Although the present study did not find that AI-associated DLBCL is more likely to present in
Jean L. Koff et al Figure 2 Survival Stratified by Autoimmune Status of Diffuse Large B-cell Lymphoma (DLBCL) Patients Treated With CHOP (Cyclophosphamide, Doxorubicin, Vincristine, Prednisone) or R-CHOP (Rituximab, CHOP). (A) Kaplan-Meier Curve of Overall Survival in DLBCL Patients Treated With CHOP or R-CHOP, Stratified by Autoimmune Status. (B) Kaplan-Meier Curve of Lymphoma-related Survival in DLBCL Patients Treated With CHOP or R-CHOP, Stratified by Autoimmune Status
A
None Sjogren's syndrome SS
RA SLE
1.00
0.80
0.60
0.40
0.20
0.00 0
1
2
3
4
5
6
7
8
9
10
11
12
13
Years from diagnosis
B
None SS Sjogren's syndrome
RA SLE
1.00
0.80
0.60
0.40
0.20
0.00 0
1
2
3
4
5
6
7
8
9
10
11
Years from diagnosis
Abbreviations: RA ¼ rheumatoid arthritis; SLE ¼ systemic lupus erythematosus; SS ¼ Sjögren syndrome.
African-American patients, it is possible that African-American patients were underrepresented in our data set.28 Because the state of Georgia is the only major southern SEER site, the SEER database might miss populations with larger proportions of African-American individuals. The large Lymphoma Epidemiology of Outcomes
cohort study is currently attempting to selectively recruit AfricanAmerican and Hispanic patients with DLBCL in centers where these patients are likely to present to overcome the limitations in analysis when such populations are underrepresented. To examine factors that might explain the racial disparity in NHL incidence
Clinical Lymphoma, Myeloma & Leukemia Month 2017
-5
Mortality and Vascular Events Among Elderly Patients With CML
6
-
patterns, Koshiol et al29 assessed immune-related conditions and the risk of developing NHL among 7999 white and 1497 AfricanAmerican hospitalized veterans. The investigators found that patients with AI disease were generally more likely to develop NHL but that the risks associated with AI conditions were similar when stratified by race.29 However, women and healthier individuals who only received outpatient care (which could be the case for patients with DLBCL) were underrepresented in this population, which likely biased these results. Given that certain AI disorders such as SLE are prevalent in African-American women and show similar patterns of an earlier age at diagnosis and increased disease severity among African-American patients,30-32 biologic factors such as host genetics and prolonged exposure to inflammatory milieu could result in an earlier diagnosis of DLBCL compared with the general population. However, this hypothesis has yet to be explored and will require populations that include younger patients and women. Although not statistically significant in our cohort, the trend toward decreased LRS in patients with concomitant SLE suggests that these patients could experience inferior outcomes compared with DLBCL patients without AI disease. DLBCL subtype as distinguished by gene expression profiling has been shown to be an important prognostic factor, with patients with the activated B-cellelike (ABC) subtype exhibiting inferior survival outcomes compared with those with germinal center B-cellelike (GCB) DLBCL.33-35 Evidence has shown that patients with AI-associated DLBCL are more likely to have the more aggressive ABC subtype. For example, a large Swedish study that included > 74,000 RA patients found that 70% of DLBCL cases arising in that cohort exhibited a noneGCB phenotype.36 However, information on DLBCL subtype is not available from the SEEReMedicare database, limiting our assessment of a possible interaction with AI disease in affecting lymphoma outcomes. Our analysis was also limited because a record of patients’ oral medications is not available in the SEEReMedicare data set. Although data on chemotherapeutic regimens for first-line treatment of DLBCL was largely extractable, most immunosuppressive therapies for AI disorders remained unavailable for our analysis. The question of whether NHL risk and outcomes for patients with systemic AI diseases such as SLE and RA stems from the immunosuppressive therapy used to treat those disorders rather than the disease activity itself remains controversial. A systematic literature review examining the treatment regimens for RA revealed no significant association of either methotrexate or azathioprine with lymphoma risk.37 Some studies have suggested that the use of tumor necrosis factor (TNF)-a inhibitors might increase NHL risk; however, the relationship is far from clearcut. A 2006 metaanalysis examining the effect of the monoclonal antibodies infliximab and adalimumab on NHL risk based on data from randomized clinical trials reported a summary odds ratio of 3.3.38 However, its inclusion of heterogeneous studies with short followup periods limited interpretation of these results. A prospective French study analyzing lymphoma risk in patients receiving antieTNF-a therapy for any inflammatory disease found that the observed two- to threefold increased risk of NHL was similar to that expected for patients with such disease.39 Also, that study found that lymphoma risk increased with the use of infliximab and adalimumab compared with etanercept, a soluble TNF-a
Clinical Lymphoma, Myeloma & Leukemia Month 2017
inhibitor. The issue is further confounded because patients with increased disease activity are more likely to be prescribed these medications, making it very difficult to confidently ascribe the increased NHL risk seen in this population to either disease severity or specific medications. Clearly, more work is needed to tease out the complicated relationships among autoimmunity, immunosuppressive therapy, and the risk of lymphoma development. Furthermore, none of these studies have addressed the relationships between AI disease severity, duration, and treatment and lymphoma outcomes. To the best of our knowledge, our study is the first to examine the outcomes of DLBCL patients with AI disease in a large population-based cohort treated with contemporary regimens. Our results suggest that most elderly patients with DLBCL and AI disease have outcomes similar to those of the general population of DLBCL patients.
Conclusion In our retrospective claims-based cohort, concomitant AI disease was uncommon and was more likely to occur in female DLBCL patients, which likely reflects the greater incidence of AI disease in women. The trend toward decreased survival in patients with SLE and DLBCL was not statistically significant; however, this might have resulted from the low number of patients in this group in our study. The possibility of lower LRS for these patients should be explored in future studies, especially those including younger patients, to capture larger numbers of patients with AIassociated DLBCL and further define the characteristics of this group.
Clinical Practice Points Little is known about the demographic data or clinical outcomes
of those with DLBCL that arises in the setting of AI disease. Given that inflammation and chronic self-antigen stimulation in
AI disease represent specific pathways that could promote lymphoma development, AI-associated lymphomas could constitute a distinct subset within DLBCL that exhibits characteristic clinical and biologic behavior. We found that older DLBCL patients with AI disease were more likely to be female but otherwise did not differ significantly from other older DLBCL patients in demographic data, treatments, and clinical outcomes. The observed trend toward decreased survival for patients with SLE and DLBCL was not statistically significant; however, the lack of significance might have resulted from the low number of patients in this group. Future studies should incorporate analyses of younger patients to capture larger numbers of individuals with AI-associated DLBCL and further define the characteristics of this group.
Disclosure The authors have stated that they have no conflicts of interest.
References 1. Ferlay J, Soerjomataram I, Ervik M, et al. GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 11 [Internet]. Lyon, France: International Agency for Research on Cancer; 2013. Available at: http://globocan. iarc.fr. Accessed: May 1, 2017.
Jean L. Koff et al 2. Swerdlow SH, International Agency for Research on Cancer and World Health Organization. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed. Lyon, France: International Agency for Research on Cancer; 2008. 3. Teras LR, DeSantis CE, Cerhan JR, et al. US lymphoid malignancy statistics by World Health Organization subtypes [e-pub ahead of print]. CA Cancer J Clin 2016. https://doi.org/10.3322/caac.21357. accessed May 1, 2017. 4. Costa LJ, Maddocks K, Epperla N, et al. Diffuse large B-cell lymphoma with primary treatment failure: ultra-high risk features and benchmarking for experimental therapies. Am J Hematol 2017; 92:161-70. 5. Flowers CR, Sinha R, Vose JM. Improving outcomes for patients with diffuse large B-cell lymphoma. CA Cancer J Clin 2010; 60:393-408. 6. Cerhan JR, Kricker A, Paltiel O, et al. Medical history, lifestyle, family history, and occupational risk factors for diffuse large B-cell lymphoma: the InterLymph NonHodgkin Lymphoma Subtypes Project. J Natl Cancer Inst Monogr 2014; 2014:15-25. 7. Zintzaras E, Voulgarelis M, Moutsopoulos HM. The risk of lymphoma development in autoimmune diseases: a meta-analysis. Arch Intern Med 2005; 165: 2337-44. 8. Voulgarelis M, Ziakas PD, Papageorgiou A, Biampa E, Tzioufas AG, Moutsopoulos HM. Prognosis and outcome of non-Hodgkin lymphoma in primary Sjögren syndrome. Medicine (Baltimore) 2012; 91:1-9. 9. Bernatsky S, Ramsay-Goldman R, Rajan R, et al. Non-Hodgkin’s lymphoma in systemic lupus erythematosus. Ann Rheum Dis 2005; 64:1507-9. 10. Ekström Smedby K, Vajdic CM, Falster M, et al. Autoimmune disorders and risk of non-Hodgkin lymphoma subtypes: a pooled analysis within the InterLymph Consortium. Blood 2008; 111:4029-38. 11. Baecklund E, Iliadou A, Askling J, et al. Association of chronic inflammation, not its treatment, with increased lymphoma risk in rheumatoid arthritis. Arthritis Rheum 2006; 54:692-701. 12. Wang SS, Vajdic CM, Linet MS, et al. Associations of non-Hodgkin Lymphoma (NHL) risk with autoimmune conditions according to putative NHL loci. Am J Epidemiol 2015; 181:406-21. 13. National Institutes of Health, National Cancer Institute, Surveillance, Epidemiology, and End Results Program. Overview of the SEER Program, Available at: https://seer.cancer.gov/about/overview.html. Accessed: May 1, 2017. 14. National Institutes of Health, National Cancer Institute, Division of Cancer Control & Population Sciences. SEER-Medicare linked database, Available at: https://healthcaredelivery.cancer.gov/seermedicare/. Accessed: May 1, 2017. 15. Fritz A, Percy C, Jack A, et al. International Classification of Diseases for Oncology. 3rd ed, 1st revision. World Health Organization. Lyon, France: International Agency for Research on Cancer; 2013. 16. Adamo M, Dickie L, Ruhl J. SEER Program Coding and Staging Manual 2016. Bethesda, MD: National Cancer Institute; 2016. 17. Wang M, Burau KD, Fang S, Wang H, Du XL. Ethnic variations in diagnosis, treatment, socioeconomic status, and survival in a large population-based cohort of elderly patients with non-Hodgkin lymphoma. Cancer 2008; 113:3231-41. 18. Griffiths R, Gleeson M, Knopf K, Danese M. Racial differences in treatment and survival in older patients with diffuse large B-cell lymphoma (DLBCL). BMC Cancer 2010; 10:625. 19. Satram-Hoang S, Reyes C, Hoang KQ, Momin F, Skettino S. Treatment practice in the elderly patient with chronic lymphocytic leukemia-analysis of the combined SEER and Medicare database. Ann Hematol 2014; 93:1335-44. 20. Davidoff AJ, Tang M, Seal B, Edelman MJ. Chemotherapy and survival benefit in elderly patients with advanced non-small-cell lung cancer. J Clin Oncol 2010; 28: 2191-7.
21. Davidoff AJ, Zuckerman IH, Pandya N, et al. A novel approach to improve health status measurement in observational claims-based studies of cancer treatment and outcomes. J Geriatr Oncol 2013; 4:157-65. 22. Griffiths R, Mikhael J, Gleeson M, Danese M, Dreyling M. Addition of rituximab to chemotherapy alone as first-line therapy improves overall survival in elderly patients with mantle cell lymphoma. Blood 2011; 118:4808-16. 23. Griffiths RI, Gleeson ML, Mikhael J, Dreyling MH, Danese MD. Comparative effectiveness and cost of adding rituximab to first-line chemotherapy for elderly patients diagnosed with diffuse large B-cell lymphoma. Cancer 2012; 118: 6079-88. 24. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol 1992; 45:613-9. 25. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chron Dis 1987; 40:373-83. 26. Flowers CR, Shenoy PJ, Borate U, et al. Examining racial differences in diffuse large B-cell lymphoma presentation and survival. Leuk Lymphoma 2013; 54: 268-76. 27. King JK, Costenbader KH. Characteristics of patients with systemic lupus erythematosus (SLE) and non-Hodgkin’s lymphoma (NHL). Clin Rheumatol 2007; 26:1491-4. 28. Shenoy PJ, Malik N, Nooka A, et al. Racial differences in the presentation and outcomes of diffuse large B-cell lymphoma in the United States. Cancer 2011; 117: 2530-40. 29. Koshiol J, Lam TK, Gridley G, Check D, Brown LM, Landgren O. Racial differences in chronic immune stimulatory conditions and risk of non-Hodgkin’s lymphoma in veterans from the United States. J Clin Oncol 2011; 29:378-85. 30. Petri M. Epidemiology of systemic lupus erythematosus. Best Pract Res Clin Rheumatol 2002; 16:847-58. 31. Danchenko N, Satia JA, Anthony MS. Epidemiology of systemic lupus erythematosus: a comparison of worldwide disease burden. Lupus 2006; 15:308-18. 32. Pons-Estel GJ, Alarcon GS, Scofield L, Reinlib L, Cooper GS. Understanding the epidemiology and progression of systemic lupus erythematosus. Semin Arthritis Rheum 2010; 39:257-68. 33. Alizadeh AA, Eisen MB, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 2000; 403:503-11. 34. Fu K, Weisenburger DD, Choi WW, et al. Addition of rituximab to standard chemotherapy improves the survival of both the germinal center B-cell-like and non-germinal center B-cell-like subtypes of diffuse large B-cell lymphoma. J Clin Oncol 2008; 26:4587-94. 35. Rosenwald A, Wright G, Chan WC, et al. The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma. N Engl J Med 2002; 346:1937-47. 36. Baecklund E, Backlin C, Iliadou A, et al. Characteristics of diffuse large B cell lymphomas in rheumatoid arthritis. Arthritis Rheum 2006; 54:3774-81. 37. Kaiser R. Incidence of lymphoma in patients with rheumatoid arthritis: a systematic review of the literature. Clin Lymphoma Myeloma 2008; 8:87-93. 38. Bongartz T, Sutton AJ, Sweeting MJ, Buchan I, Matteson EL, Montori V. AntiTNF antibody therapy in rheumatoid arthritis and the risk of serious infections and malignancies: systematic review and meta-analysis of rare harmful effects in randomized controlled trials. JAMA 2006; 295:2275-85. 39. Mariette X, Tubach F, Bagheri H, et al. Lymphoma in patients treated with antiTNF: results of the 3-year prospective French RATIO registry. Ann Rheum Dis 2010; 69:400-8.
Clinical Lymphoma, Myeloma & Leukemia Month 2017
-7
Characterizing Autoimmune Disease-associated Diffuse Large B-cell Lymphoma in a SEEReMedicare Cohort Jean L. Koff,1 Ashish Rai,2 Christopher R. Flowers1 Abstract We used the Surveillance, Epidemiology, and End Results (SEER)eMedicare-linked database to characterize the patterns of presentation, treatment, and survival in older diffuse large B-cell lymphoma (DLBCL) patients with concomitant autoimmune disease. DLBCL patients with autoimmune disease were more likely to be female but otherwise did not differ significantly from other DLBCL patients in demographic data, treatment, or clinical outcomes. Background: Severe immune dysregulation such as seen in autoimmune (AI) disease is known to act as a significant risk factor for diffuse large B-cell lymphoma (DLBCL). However, little is known about the demographics or clinical outcomes of DLBCL that arises in the setting of AI disease. Patients and Methods: We used the Surveillance, Epidemiology, and End Results (SEER) database for patients with a diagnosis from 1999 to 2009 linked to their Medicare claims data through 2011 to characterize the presentation, treatment, and survival patterns in DLBCL patients, including those with rheumatoid arthritis, systemic lupus erythematosus (SLE), Sjögren syndrome, and other B-cellemediated AI diseases. We examined the baseline clinical characteristics for patients with B-celle mediated AI disease, plotted the overall survival and lymphoma-related survival (LRS) for these groups, and compared the median survival times. Results: Patients with DLBCL and AI disease were more commonly female. However, patients with DLBCL and rheumatoid arthritis, SLE, Sjögren syndrome, or other B-cell AI diseases did not differ from other DLBCL patients in any other baseline presenting features and received similar first-line treatment. A trend toward decreased LRS was seen in patients with SLE and DLBCL compared with all other groups, but this difference was not statistically significant in this cohort. Conclusion: In the present retrospective claims-based cohort of older patients with DLBCL, concomitant AI disease was uncommon and was more likely to occur in female DLBCL patients, which likely reflects the greater incidence of AI disease in women. The possibility of lower LRS for SLE patients should be explored in future studies. Clinical Lymphoma, Myeloma & Leukemia, Vol. -, No. -, --- ª 2017 Elsevier Inc. All rights reserved. Keywords: DLBCL, Non-Hodgkin lymphoma, Rheumatoid arthritis, Sjögren syndrome, Systemic lupus erythematosus
Introduction Non-Hodgkin lymphoma (NHL) is the most common hematologic malignancy worldwide.1 Its most frequently diagnosed subtype, diffuse large B-cell lymphoma (DLBCL), accounts for w25% to 30% NHL cases in Western countries.2,3 Importantly, although 1 Department of Hematology and Medical Oncology, Winship Cancer Institute of Emory University, Atlanta, GA 2 American Cancer Society, Atlanta, GA
Submitted: Jun 20, 2017; Revised: Nov 17, 2017; Accepted: Nov 27, 2017 Address for correspondence: Jean L. Koff, MD, Department of Hematology and Medical Oncology, Winship Cancer Institute of Emory University, 1365 Clifton Road, Building B, Suite 4300, Atlanta, GA 30322 Fax: (404) 778-3366; e-mail contact: [email protected]
2152-2650/$ - see frontmatter ª 2017 Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.clml.2017.11.009
DLBCLs appear histologically similar, they demonstrate a wide degree of genetic and clinical heterogeneity. DLBCL can be curable; however, nearly 50% of patients will eventually relapse, with subsequent dismal prognosis.4,5 Thus, substantial research efforts have been focused on identifying the biologic and clinical features that underlie these stark differences in outcomes with the hope that targeted therapeutic approaches could abrogate such disparities. Severe immune dysregulation such as that seen with human immunodeficiency virus infection, immunosuppression after solid organ transplantation, and autoimmune (AI) disease is known to act as a major risk factor for NHL. The InterLymph Subtypes Project pooled cases and controls to provide well-powered comparisons of risk factors for specific NHL subtypes, including DLBCL. Multivariate analysis of 4667 DLBCL cases and 22,639
Clinical Lymphoma, Myeloma & Leukemia Month 2017
-1
Mortality and Vascular Events Among Elderly Patients With CML controls showed that B-celleactivating AI diseases in general (odds ratio, 2.4; 95% confidence interval, 1.8-3.1), and systemic lupus erythematosus (SLE) and Sjögren syndrome (SS) in particular, were the most strongly associated with increased DLBCL risk after controlling for all other risk factors.6 These findings mirror those of large cohort studies of patients with AI disease. SS has long been known to carry the greatest risk of NHL (relative risk, 4-40 times that of the general population), most of which are mucosaassociated lymphoid tissue lymphomas.7 However, DLBCL cases comprised about 15% of SS-associated lymphoma in a 584-patient cohort.8 Similarly, a 2005 cohort study involving 9547 SLE patients found that aggressive subtypes predominated among the NHL diagnoses, with DLBCL constituting more than one half of the cases for which subtype was specified.9 The significant heterogeneity in the studies of NHL risk in patients with rheumatoid arthritis (RA) preempted the InterLymph group from performing focused analyses of NHL subtype risk in that population.10 However, a large Swedish study that included > 74,000 RA patients found that patients with greater indexes of inflammatory activity exhibited an increased risk of NHL in general and DLBCL in particular (48% of cases).11 Given that the processes of inflammation and chronic selfantigen stimulation that define AI diseases represent specific pathways that could promote lymphoma development, it is possible that AI-associated lymphomas are a distinct subset within DLBCL exhibiting characteristic clinical and biologic behavior. However, little is known about the demographic data or clinical outcomes of DLBCL that arises in the setting of AI disease. We examined the Surveillance, Epidemiology, and End Results (SEER)eMedicarelinked database to determine the frequency of common B-cell AI diseases among older patients with DLBCL and to characterize the patterns of presentation, treatment, and survival for DLBCL patients with concomitant AI disease.
Patients and Methods Data Source
2
-
We used the National Cancer Institute (NCI) SEER database for patients with a diagnosis from 1999 to 2009 linked to their Medicare claims data through 2011 to characterize the presentation, treatment, and survival patterns in patients with DLBCL, including those with RA, SLE, SS, and other B-cellemediated AI diseases as defined by the InterLymph criteria (AI hemolytic anemia, Hashimoto’s thyroiditis/hypothyroidism, myasthenia gravis, and pernicious anemia12). The SEER program reports data on cancer incidence and survival collected from US registries, covering about 28% of the population as of 2016.13 The collected data include patient demographics, tumor pathology, disease stage, primary tumor site, first-line treatment, and dates of diagnosis and death. Linking SEER with the Medicare claims data allows for the identification of concomitant health conditions and specific treatments received by elderly cancer patients. Among individuals > 65 years, 97% are Medicare-eligible, and 93% of those listed in SEER are linked to the Medicare enrollment file.14 Because this database does not include patient identifiers, our study did not require approval from an institutional review board. However, a data use agreement was signed before initiation of the study.
Clinical Lymphoma, Myeloma & Leukemia Month 2017
Eligibility Criteria Patients were considered eligible for analysis if they had received a diagnosis of DLBCL from January 1, 2002 to December 31, 2009, had linked Medicaid claims data available 2011, and were aged 66 years at diagnosis. The minimum required age was 66 years to ensure that patients had been enrolled in Medicare for 12 months before the diagnosis. Cases were identified using the World Health Organization International Classification of Diseases for Oncology, 3rd edition, histology codes 9680 and 9684.15 The following International Classification of Diseases, 9th revision, Clinical Modification codes were used to identify the presence of concomitant AI disease: SLE, 710.0; RA, 714.0 to 714.3; AI hemolytic anemia, 283.0; myasthenia gravis, 358.00 and 358.01; pernicious anemia, 281.0; and SS: 710.2. The exclusion criteria are shown in Figure 1.
Patient Characteristics Patients were stratified into groups by AI disease (none coded, SLE, SS, RA, or other B-celleactivating AI disease) as identified in Medicare claims. Self-reported race was categorized as white, African-American, or other; using SEER data, the “other” category includes individuals of Asian, Native American, Pacific Islander, or Alaskan Native ancestry.16 The SEEReMedicare database uses census tract information (eg, percentage of residents living in poverty and percentage with only a high school education) from the 2000 US Census as a surrogate for socioeconomic status, as described in other SEEReMedicare studies.17-19 Other demographic variables analyzed in the present study included sex, marital status, and type of geographic area (less urban or rural, urban, or metropolitan). In terms of disease status, patients were classified with regard to the following: Ann Arbor stage (I/II, III/IV, or unknown), primary disease site (nodal vs. extranodal), presence of B symptoms, performance status, NCI comorbidity index score (0, 1, or 2), and year of diagnosis. Performance status was classified as poor if a patient had claims for any of the following: hospice, home health agency, skilled nursing facility, oxygen, or wheelchair or related supplies. Such claims-based measures of performance status have been used in other cancer studies.20-23 The NCI comorbidity index scores were calculated using the Deyo adaptation of the Charlson comorbidity index to identify the 15 noncancer comorbidities included in the Charlson comorbidity index from the Medicare claims.24,25
Treatment and Mortality Classification We determined the initial management strategies using the Medicare claims made within 6 months of diagnosis; if no treatment was documented within that period, management was categorized as “observation.” The SEEReMedicare data set does not include information regarding the receipt of oral medications without an intravenous equivalent. Thus, patients with claims for cyclophosphamide, doxorubicin, and vincristine were categorized as having received CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone), and those with claims for cyclophosphamide and vincristine were classified as having received CVP (cyclophosphamide, vincristine, prednisolone). Patients with those same claims who also received rituximab were classified as having received R-CHOP and R-CVP, respectively.
Jean L. Koff et al Figure 1 Selection of the Study Cohort
All DLBCL patients in the SEER-Medicare database diagnosed from 1999 through 2009 n=33,886
Less patients with unknown age at diagnosis or diagnosis after death or SEER and Medicare reporting different dates of death n=469
DLBCL patients with known age at diagnosis n=33,417
DLBLC patients in patients aged 66 and older n=24,387
Less patients with age at diagnosis less than 66 years n=9,030 Less patients with interrupted Medicare coverage and/or HMO enrolment 12 months before and 6 months after diagnosis n=12,930
DLBCL patients with adequate Medicare coverage n=11,457
Less patients that were not histologically confirmed n=512
Adequately diagnosed DLBCL patients n=10,945
Less patients with date of initiation of therapy more than 45 days before date of diagnosis n=506
Eligible DLBCL patients initially observed or treated n=10,439 Less patients initially treated with regimens other than R-CHOP, R-CVP, CHOP, or CVP n=4,515 Study cohort of DLBCL patients n=5,924 DLBCL patients with rheumatoid arthritis n=155 DLBCL patients with systemic lupus erythematosus n=25
DLBCL patients with any B-cell autoimmune disease n=270
DLBCL patients with Sjogren's syndrome n=18
Abbreviations: CHOP ¼ cyclophosphamide, doxorubicin, vincristine, prednisone; CVP ¼ cyclophosphamide, vincristine, prednisolone; DLBCL ¼ diffuse large B-cell lymphoma; HMO ¼ health maintenance organization; R ¼ rituximab; SEER ¼ Surveillance, Epidemiology, and End Results.
We used the SEER date and cause of death to determine the mortality classification. Because SEER only reports diagnoses by month and year, the date of diagnosis for our survival analyses was designated as the 15th day of the reported month. Patients were followed up until death, enrollment in a health maintenance organization, or last date of available Medicare claims. We examined 2 different survival endpoints: overall survival (OS), measured from the date of diagnosis until death, censored at the last follow-up; and lymphoma-related survival (LRS), measured from the date of diagnosis until death from lymphoma, censored at death from other causes or the last follow-up.
Statistical Analysis Patients without concomitant AI disease were compared to patients with SLE, SS, RA, or any B-cellemediated AI disease using c2 tests. Multivariable logistic regression models were used to investigate the relationships between patient characteristics and treatment regimen, and odds ratios with 95% confidence intervals (CIs) were calculated. Multivariable regression models were adjusted for the following demographic and clinical variables: sex, race, marital status, percentage in the census tract living in poverty, percentage in the census tract with only a high school education, type of geographic area, disease stage, primary disease site, presence of B symptoms, NCI comorbidity index score, performance status,
and year of diagnosis. We used the Hosmer-Lemeshow test to assess the goodness of fit of the logistic regression model, which was found to fit the data well. Kaplan-Meier curves were constructed to examine the effect of concurrent AI disease diagnosis on OS and LRS. Cox proportional hazards models were adjusted for the same variables described. We tested the global proportional hazards assumption with the Wald test, and proportional hazards assumptions for individual covariates were tested by assessing Schoenfeld residuals. No violations were detected. Sensitivity analyses were performed using propensity score methods to adjust for imbalances in observable covariates between treatment groups. We set a at 0.05 to determine the statistical significance, and all P values were 2-sided. Data were analyzed using SAS, version 9.4 (SAS Institute, Cary, NC), and Stata, version 13 (StataCorp LP, College Station, TX).
Results We identified 5926 patients with DLBCL, of whom 270 had B-cellemediated AI disease. The patient characteristics are summarized in Table 1. The SEEReMedicare data use agreement stipulates that patient data with numbers < 11 should not be directly reported or be derivable with more precision than “n < 11,” which limits the presentation of baseline characteristics for patients with concurrent DLBCL and either RA (n ¼ 155), SLE (n ¼ 25),
Clinical Lymphoma, Myeloma & Leukemia Month 2017
-3
Mortality and Vascular Events Among Elderly Patients With CML Table 1 Demographic and Disease Characteristics of DLBCL Patients Stratified by AI Disease in SEEReMedicare Database, 1999-2009 Characteristic Total patients
All Patients
B-cell AI Disease
5924
270
Age, y Median
75
75
71-80
71-79
Male
2811 (47)
79 (29)
Female
3113 (53)
191 (71)
IQR Sex
Table 2 Cox Regression Analysis of Overall Survival Stratified by Stage AI Disease
All Stages
Stage I/II
Stage III/IV
None RA SLE SS Other B-cell AI disease
Reference 0.86 (0.69-1.08) 1.17 (0.70-1.95) 1.23 (0.68-2.23) 1.22 (0.90-1.64)
Reference 0.82 (0.60-1.12) 0.97 (0.40-2.37) 1.07 (0.44-2.63) 1.21 (0.82-1.79)
Reference 0.83 (0.58-1.19) 1.22 (0.60-2.48) 1.59 (0.65-3.87) 1.25 (0.77-2.04)
Data presented as hazard ratio (95% confidence interval). Abbreviations: AI ¼ autoimmune; RA ¼ rheumatoid arthritis; SLE ¼ systemic lupus erythematosus; SS ¼ Sjögren syndrome.
Race White
5341 (90)
243 (90)
African American
185 (3)
NA
Other
398 (7)
NA
Stage I/II
3224 (54)
147 (54)
III/IV
2290 (39)
107 (40)
410 (7)
16 (6)
Nodal
3820 (64)
168 (62)
Extranodal
2104 (36)
102 (38)
Absent
4835 (82)
211 (78)
Present
1089 (18)
59 (22)
0
3738 (63)
48 (18)
1
1422 (24)
119 (44)
764 (13)
103 (38)
Unknown Nodal status
Poor performance status
NCI comorbidity index
>2 Frontline management CHOP
1025 (17)
47 (17)
R-CHOP
3346 (57)
157 (58)
122 (2)
NA
CVP R-CVP No recorded chemotherapy
345 (6)
NA
1086 (18)
42 (16)
OS for R-CHOP and CHOP patients Median
8.0
7.1
95% CI
7.7-8.4
5.0-8.7
R-CHOP, R-CVP, CHOP, or CVP), 40% had missing data for the agents used, 7% contained both rituximab and an anthracycline, 24% included rituximab but no anthracycline, 3% included an anthracycline without rituximab, and 7% contained neither rituximab nor an anthracycline; 19% of the patients had received radiation alone. The total number of chemotherapy cycles received did not vary significantly by AI disease status, with 54% of all patients receiving 4 cycles compared with 58% of RA, 68% of SLE, and 60% of SS patients. Similarly, 31% of all patients received 6 cycles compared with 38% of RA, 45% of SLE, and 33% of SS patients. The median OS for patients with AI disease did not differ significantly from that seen for patients without AI disease. The median OS was 6.34 years (95% CI, 4.55-8.60 years) for RA patients, 3.23 years (95% CI, 2.14-11.97 years) for SLE patients, 5.51 years (95% CI, 2.42 years to not reached) for SS patients, and 5.97 years (95% CI, 4.45-7.78 years) for patients with any Bcellemediated AI disease compared with 6.96 years (95% CI, 6.697.37 years) for patients without AI disease. A trend toward decreased LRS was seen in patients with SLE and DLBCL compared with other groups; however, this difference was not statistically significant in this cohort (Table 2 and Figure 2). The median LRS was not reached for any group. Because few differences were found in the baseline characteristics among the groups, traditional multivariable regression models are reported rather than propensity-matched models.
Discussion Data presented as n (%). Abbreviations: AI ¼ autoimmune; CHOP ¼ cyclophosphamide, doxorubicin, vincristine, prednisone; CI ¼ confidence interval; CVP ¼ cyclophosphamide, vincristine, prednisone; IQR ¼ interquartile range; NA ¼ not available (data for which n < 11 limited direct reporting per the SEEReMedicare data use agreement); NCI ¼ National Cancer Institute; OS ¼ overall survival; R ¼ rituximab.
4
-
or SS (n ¼ 18). With the exception that patients with concomitant DLBCL and AI disease were more commonly female, patients with DLBCL and RA, SLE, SS, or other B-cell AI diseases had baseline presenting features similar to those of other DLBCL patients and received similar first-line treatment. More specifically, no statistically significant differences were found in the receipt of either anthracycline-containing regimens (74% of all patients, 78% of RA, 80% of SLE, 83% of SS) or rituximab (63% of all patients, 66% of RA, 60% of SLE, 72% of SS). Among the treatment regimens excluded from analysis (ie, regimens other than chemotherapy with
Clinical Lymphoma, Myeloma & Leukemia Month 2017
Our results serve as an initial characterization of AI-associated DLBCL in older patients. By virtue of the use of Medicare claims to capture data on AI diagnosis and cancer treatments, our analysis was limited to patients > 65 years. Thus, we would expect to capture more cases of de novo and AI-associated DLBCL by analysis of data sets that include younger patients. Our group has previously shown that African-American DLBCL patients present at a younger age and with more aggressive disease compared with white patients.26 Reported data have suggested that such a relationship might exist for patients with AI disease as well. For example, the median age of SLE patients at NHL diagnosis was 49 years in 1 study of 11 cases27 and 57 years in a 2005 study that included 42 cases.9 Also, a 2012 study of SS-associated NHL reported a median age at DLBCL diagnosis of 66.5 years.8 Although the present study did not find that AI-associated DLBCL is more likely to present in
Jean L. Koff et al Figure 2 Survival Stratified by Autoimmune Status of Diffuse Large B-cell Lymphoma (DLBCL) Patients Treated With CHOP (Cyclophosphamide, Doxorubicin, Vincristine, Prednisone) or R-CHOP (Rituximab, CHOP). (A) Kaplan-Meier Curve of Overall Survival in DLBCL Patients Treated With CHOP or R-CHOP, Stratified by Autoimmune Status. (B) Kaplan-Meier Curve of Lymphoma-related Survival in DLBCL Patients Treated With CHOP or R-CHOP, Stratified by Autoimmune Status
A
None Sjogren's syndrome SS
RA SLE
1.00
0.80
0.60
0.40
0.20
0.00 0
1
2
3
4
5
6
7
8
9
10
11
12
13
Years from diagnosis
B
None SS Sjogren's syndrome
RA SLE
1.00
0.80
0.60
0.40
0.20
0.00 0
1
2
3
4
5
6
7
8
9
10
11
Years from diagnosis
Abbreviations: RA ¼ rheumatoid arthritis; SLE ¼ systemic lupus erythematosus; SS ¼ Sjögren syndrome.
African-American patients, it is possible that African-American patients were underrepresented in our data set.28 Because the state of Georgia is the only major southern SEER site, the SEER database might miss populations with larger proportions of African-American individuals. The large Lymphoma Epidemiology of Outcomes
cohort study is currently attempting to selectively recruit AfricanAmerican and Hispanic patients with DLBCL in centers where these patients are likely to present to overcome the limitations in analysis when such populations are underrepresented. To examine factors that might explain the racial disparity in NHL incidence
Clinical Lymphoma, Myeloma & Leukemia Month 2017
-5
Mortality and Vascular Events Among Elderly Patients With CML
6
-
patterns, Koshiol et al29 assessed immune-related conditions and the risk of developing NHL among 7999 white and 1497 AfricanAmerican hospitalized veterans. The investigators found that patients with AI disease were generally more likely to develop NHL but that the risks associated with AI conditions were similar when stratified by race.29 However, women and healthier individuals who only received outpatient care (which could be the case for patients with DLBCL) were underrepresented in this population, which likely biased these results. Given that certain AI disorders such as SLE are prevalent in African-American women and show similar patterns of an earlier age at diagnosis and increased disease severity among African-American patients,30-32 biologic factors such as host genetics and prolonged exposure to inflammatory milieu could result in an earlier diagnosis of DLBCL compared with the general population. However, this hypothesis has yet to be explored and will require populations that include younger patients and women. Although not statistically significant in our cohort, the trend toward decreased LRS in patients with concomitant SLE suggests that these patients could experience inferior outcomes compared with DLBCL patients without AI disease. DLBCL subtype as distinguished by gene expression profiling has been shown to be an important prognostic factor, with patients with the activated B-cellelike (ABC) subtype exhibiting inferior survival outcomes compared with those with germinal center B-cellelike (GCB) DLBCL.33-35 Evidence has shown that patients with AI-associated DLBCL are more likely to have the more aggressive ABC subtype. For example, a large Swedish study that included > 74,000 RA patients found that 70% of DLBCL cases arising in that cohort exhibited a noneGCB phenotype.36 However, information on DLBCL subtype is not available from the SEEReMedicare database, limiting our assessment of a possible interaction with AI disease in affecting lymphoma outcomes. Our analysis was also limited because a record of patients’ oral medications is not available in the SEEReMedicare data set. Although data on chemotherapeutic regimens for first-line treatment of DLBCL was largely extractable, most immunosuppressive therapies for AI disorders remained unavailable for our analysis. The question of whether NHL risk and outcomes for patients with systemic AI diseases such as SLE and RA stems from the immunosuppressive therapy used to treat those disorders rather than the disease activity itself remains controversial. A systematic literature review examining the treatment regimens for RA revealed no significant association of either methotrexate or azathioprine with lymphoma risk.37 Some studies have suggested that the use of tumor necrosis factor (TNF)-a inhibitors might increase NHL risk; however, the relationship is far from clearcut. A 2006 metaanalysis examining the effect of the monoclonal antibodies infliximab and adalimumab on NHL risk based on data from randomized clinical trials reported a summary odds ratio of 3.3.38 However, its inclusion of heterogeneous studies with short followup periods limited interpretation of these results. A prospective French study analyzing lymphoma risk in patients receiving antieTNF-a therapy for any inflammatory disease found that the observed two- to threefold increased risk of NHL was similar to that expected for patients with such disease.39 Also, that study found that lymphoma risk increased with the use of infliximab and adalimumab compared with etanercept, a soluble TNF-a
Clinical Lymphoma, Myeloma & Leukemia Month 2017
inhibitor. The issue is further confounded because patients with increased disease activity are more likely to be prescribed these medications, making it very difficult to confidently ascribe the increased NHL risk seen in this population to either disease severity or specific medications. Clearly, more work is needed to tease out the complicated relationships among autoimmunity, immunosuppressive therapy, and the risk of lymphoma development. Furthermore, none of these studies have addressed the relationships between AI disease severity, duration, and treatment and lymphoma outcomes. To the best of our knowledge, our study is the first to examine the outcomes of DLBCL patients with AI disease in a large population-based cohort treated with contemporary regimens. Our results suggest that most elderly patients with DLBCL and AI disease have outcomes similar to those of the general population of DLBCL patients.
Conclusion In our retrospective claims-based cohort, concomitant AI disease was uncommon and was more likely to occur in female DLBCL patients, which likely reflects the greater incidence of AI disease in women. The trend toward decreased survival in patients with SLE and DLBCL was not statistically significant; however, this might have resulted from the low number of patients in this group in our study. The possibility of lower LRS for these patients should be explored in future studies, especially those including younger patients, to capture larger numbers of patients with AIassociated DLBCL and further define the characteristics of this group.
Clinical Practice Points Little is known about the demographic data or clinical outcomes
of those with DLBCL that arises in the setting of AI disease. Given that inflammation and chronic self-antigen stimulation in
AI disease represent specific pathways that could promote lymphoma development, AI-associated lymphomas could constitute a distinct subset within DLBCL that exhibits characteristic clinical and biologic behavior. We found that older DLBCL patients with AI disease were more likely to be female but otherwise did not differ significantly from other older DLBCL patients in demographic data, treatments, and clinical outcomes. The observed trend toward decreased survival for patients with SLE and DLBCL was not statistically significant; however, the lack of significance might have resulted from the low number of patients in this group. Future studies should incorporate analyses of younger patients to capture larger numbers of individuals with AI-associated DLBCL and further define the characteristics of this group.
Disclosure The authors have stated that they have no conflicts of interest.
References 1. Ferlay J, Soerjomataram I, Ervik M, et al. GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 11 [Internet]. Lyon, France: International Agency for Research on Cancer; 2013. Available at: http://globocan. iarc.fr. Accessed: May 1, 2017.
Jean L. Koff et al 2. Swerdlow SH, International Agency for Research on Cancer and World Health Organization. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed. Lyon, France: International Agency for Research on Cancer; 2008. 3. Teras LR, DeSantis CE, Cerhan JR, et al. US lymphoid malignancy statistics by World Health Organization subtypes [e-pub ahead of print]. CA Cancer J Clin 2016. https://doi.org/10.3322/caac.21357. accessed May 1, 2017. 4. Costa LJ, Maddocks K, Epperla N, et al. Diffuse large B-cell lymphoma with primary treatment failure: ultra-high risk features and benchmarking for experimental therapies. Am J Hematol 2017; 92:161-70. 5. Flowers CR, Sinha R, Vose JM. Improving outcomes for patients with diffuse large B-cell lymphoma. CA Cancer J Clin 2010; 60:393-408. 6. Cerhan JR, Kricker A, Paltiel O, et al. Medical history, lifestyle, family history, and occupational risk factors for diffuse large B-cell lymphoma: the InterLymph NonHodgkin Lymphoma Subtypes Project. J Natl Cancer Inst Monogr 2014; 2014:15-25. 7. Zintzaras E, Voulgarelis M, Moutsopoulos HM. The risk of lymphoma development in autoimmune diseases: a meta-analysis. Arch Intern Med 2005; 165: 2337-44. 8. Voulgarelis M, Ziakas PD, Papageorgiou A, Biampa E, Tzioufas AG, Moutsopoulos HM. Prognosis and outcome of non-Hodgkin lymphoma in primary Sjögren syndrome. Medicine (Baltimore) 2012; 91:1-9. 9. Bernatsky S, Ramsay-Goldman R, Rajan R, et al. Non-Hodgkin’s lymphoma in systemic lupus erythematosus. Ann Rheum Dis 2005; 64:1507-9. 10. Ekström Smedby K, Vajdic CM, Falster M, et al. Autoimmune disorders and risk of non-Hodgkin lymphoma subtypes: a pooled analysis within the InterLymph Consortium. Blood 2008; 111:4029-38. 11. Baecklund E, Iliadou A, Askling J, et al. Association of chronic inflammation, not its treatment, with increased lymphoma risk in rheumatoid arthritis. Arthritis Rheum 2006; 54:692-701. 12. Wang SS, Vajdic CM, Linet MS, et al. Associations of non-Hodgkin Lymphoma (NHL) risk with autoimmune conditions according to putative NHL loci. Am J Epidemiol 2015; 181:406-21. 13. National Institutes of Health, National Cancer Institute, Surveillance, Epidemiology, and End Results Program. Overview of the SEER Program, Available at: https://seer.cancer.gov/about/overview.html. Accessed: May 1, 2017. 14. National Institutes of Health, National Cancer Institute, Division of Cancer Control & Population Sciences. SEER-Medicare linked database, Available at: https://healthcaredelivery.cancer.gov/seermedicare/. Accessed: May 1, 2017. 15. Fritz A, Percy C, Jack A, et al. International Classification of Diseases for Oncology. 3rd ed, 1st revision. World Health Organization. Lyon, France: International Agency for Research on Cancer; 2013. 16. Adamo M, Dickie L, Ruhl J. SEER Program Coding and Staging Manual 2016. Bethesda, MD: National Cancer Institute; 2016. 17. Wang M, Burau KD, Fang S, Wang H, Du XL. Ethnic variations in diagnosis, treatment, socioeconomic status, and survival in a large population-based cohort of elderly patients with non-Hodgkin lymphoma. Cancer 2008; 113:3231-41. 18. Griffiths R, Gleeson M, Knopf K, Danese M. Racial differences in treatment and survival in older patients with diffuse large B-cell lymphoma (DLBCL). BMC Cancer 2010; 10:625. 19. Satram-Hoang S, Reyes C, Hoang KQ, Momin F, Skettino S. Treatment practice in the elderly patient with chronic lymphocytic leukemia-analysis of the combined SEER and Medicare database. Ann Hematol 2014; 93:1335-44. 20. Davidoff AJ, Tang M, Seal B, Edelman MJ. Chemotherapy and survival benefit in elderly patients with advanced non-small-cell lung cancer. J Clin Oncol 2010; 28: 2191-7.
21. Davidoff AJ, Zuckerman IH, Pandya N, et al. A novel approach to improve health status measurement in observational claims-based studies of cancer treatment and outcomes. J Geriatr Oncol 2013; 4:157-65. 22. Griffiths R, Mikhael J, Gleeson M, Danese M, Dreyling M. Addition of rituximab to chemotherapy alone as first-line therapy improves overall survival in elderly patients with mantle cell lymphoma. Blood 2011; 118:4808-16. 23. Griffiths RI, Gleeson ML, Mikhael J, Dreyling MH, Danese MD. Comparative effectiveness and cost of adding rituximab to first-line chemotherapy for elderly patients diagnosed with diffuse large B-cell lymphoma. Cancer 2012; 118: 6079-88. 24. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol 1992; 45:613-9. 25. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chron Dis 1987; 40:373-83. 26. Flowers CR, Shenoy PJ, Borate U, et al. Examining racial differences in diffuse large B-cell lymphoma presentation and survival. Leuk Lymphoma 2013; 54: 268-76. 27. King JK, Costenbader KH. Characteristics of patients with systemic lupus erythematosus (SLE) and non-Hodgkin’s lymphoma (NHL). Clin Rheumatol 2007; 26:1491-4. 28. Shenoy PJ, Malik N, Nooka A, et al. Racial differences in the presentation and outcomes of diffuse large B-cell lymphoma in the United States. Cancer 2011; 117: 2530-40. 29. Koshiol J, Lam TK, Gridley G, Check D, Brown LM, Landgren O. Racial differences in chronic immune stimulatory conditions and risk of non-Hodgkin’s lymphoma in veterans from the United States. J Clin Oncol 2011; 29:378-85. 30. Petri M. Epidemiology of systemic lupus erythematosus. Best Pract Res Clin Rheumatol 2002; 16:847-58. 31. Danchenko N, Satia JA, Anthony MS. Epidemiology of systemic lupus erythematosus: a comparison of worldwide disease burden. Lupus 2006; 15:308-18. 32. Pons-Estel GJ, Alarcon GS, Scofield L, Reinlib L, Cooper GS. Understanding the epidemiology and progression of systemic lupus erythematosus. Semin Arthritis Rheum 2010; 39:257-68. 33. Alizadeh AA, Eisen MB, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 2000; 403:503-11. 34. Fu K, Weisenburger DD, Choi WW, et al. Addition of rituximab to standard chemotherapy improves the survival of both the germinal center B-cell-like and non-germinal center B-cell-like subtypes of diffuse large B-cell lymphoma. J Clin Oncol 2008; 26:4587-94. 35. Rosenwald A, Wright G, Chan WC, et al. The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma. N Engl J Med 2002; 346:1937-47. 36. Baecklund E, Backlin C, Iliadou A, et al. Characteristics of diffuse large B cell lymphomas in rheumatoid arthritis. Arthritis Rheum 2006; 54:3774-81. 37. Kaiser R. Incidence of lymphoma in patients with rheumatoid arthritis: a systematic review of the literature. Clin Lymphoma Myeloma 2008; 8:87-93. 38. Bongartz T, Sutton AJ, Sweeting MJ, Buchan I, Matteson EL, Montori V. AntiTNF antibody therapy in rheumatoid arthritis and the risk of serious infections and malignancies: systematic review and meta-analysis of rare harmful effects in randomized controlled trials. JAMA 2006; 295:2275-85. 39. Mariette X, Tubach F, Bagheri H, et al. Lymphoma in patients treated with antiTNF: results of the 3-year prospective French RATIO registry. Ann Rheum Dis 2010; 69:400-8.
Clinical Lymphoma, Myeloma & Leukemia Month 2017
-7