Outcomes of Chemotherapy Treatment in Elderly Patients with Hodgkin Lymphoma

Introduction: Hodgkin lymphoma (HL) is a common lymphoid malignancy accounting for approximately 10 percent of all lymphomas. In the United States (US), the incidence of HL has a bimodal distribution curve with peaks at age 20 years and a second peak at approximately age 65 years. Age of diagnosis is an important prognostic marker in HL with younger patients having superior outcomes. Patients diagnosed at greater than age 60 years are considered elderly. Prior analyses of HL treatment in elderly patients have revealed worse outcomes in this population with reported 5-year progression free survival (PFS) and overall survival (OS) from 30-45% and 40-60% ranges, respectively (Evens, Blood 2012). The most common first line treatment for HL is doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD). Elderly patients are thought to have higher risk of bleomycin toxicity and worse outcomes with treatment in comparison to younger patients. Omission of bleomycin or doxorubicin from chemotherapy regimens administered to elderly patients is not uncommon. This study evaluated treatment outcomes of HL in an elderly U.S. Veteran population.

Methods: The VHA database was used to identify patients diagnosed with HL using ICD-03 codes C000-C8099, 9651, 9652, 9653-9655, 9663-9667, 9650, 9661-9662, and 9659. Data was available from the Veterans Affairs Central Cancer Registry (VACCR) from October 1^st, 2001-December 31^st 2013. Data was obtained on patient demographics, weight, height, other clinical characteristics, date of diagnosis and HL directed treatments. Chemotherapeutic agents patients received were collected from the VACCR data and missing data was abstracted using chart review. Patients who received at least one dose of any chemotherapy for HL were considered to have received chemotherapy. Of patients who received chemotherapy, those who received at least one dose of bleomycin or doxorubicin were included in the bleomycin and doxorubicin groups respectively. Patients less than age 60 years were excluded. Patients greater than age 60 years were stratified by age into age groups 61-69 and ≥70 years. Overall survival was measured from date of diagnosis to date of death. Kaplan-Meier method for overall survival was used to evaluate each treatment group. Cox proportional hazard model was used to evaluate risk of death after adjusting for age (61-69 years used as reference) and chemotherapy used (bleomycin or doxorubicin).

Results: There were 1,135 unique HL patients [632 (56%) patients ≤ age 60 years, 503 (44%) patients >age 60 years] identified in the VHA database. For patients age years >60, 389 patients (77%) received chemotherapy; with 226 (58%) patients age years 61-69 and 163 (42%) age years ≥70. Bleomycin was administered to 177 (78%) patients age years 61-69 and 106 (65%) patients age years ≥70 (p=0.037). Doxorubicin was administered to 198 (88%) patients age years 61-69 and to 122 (75%) patients age years ≥70 (p=0.0012). Cox proportional hazards analysis demonstrated that patients ≥70 years had a worse overall survival (HR = 1.71; 95% CI 1.276-2.293) in comparison patients age 61-69 years. After adjusting for age and chemotherapy, treatment with doxorubicin (HR 0.67; 95% CI 0.416-1.084) or bleomycin (HR 0.77; 95% CI 0.512-1.192) was associated with an improved survival.

Conclusions: This retrospective analysis represents the largest cohort of elderly HL patients with detailed treatment information reported to date. Advanced age (age ≥70 years) was confirmed to be a poor prognostic factor. The chemotherapy agents used in elderly patients also differed by age with less doxorubicin and bleomycin administered to patients age ≥70 years in comparison to those age years 61-69. While not statistically significant, this study demonstrated improved survival in patients receiving bleomycin or doxorubicin regardless of age. Further analyses adjusting for patient comorbidities, evaluating for treatment related toxicity (i.e. pulmonary toxicity) and accounting for cause specific mortality are pending.