Abstract 4287


Greater than 50% of newly diagnosed non-Hodgkin lymphoma (NHL) cases are ≥ age 60, and the numbers of older patients with NHL will grow as the population ages. NHL treatment may significantly impact health related quality of life (HRQOL) in older patients and thereby have long-term adverse physical and mental consequences. One recent evaluation of HRQOL following a cancer diagnosis among patients age '65 indicated a significant decrease following any NHL diagnosis, particularly in the physical health component. However, NHL is a collection of distinct disease entities, with widely varied clinical course. Diffuse large B-cell lymphoma (DLBCL) is a prevalent subtype often treated with curative intent. In this analysis, we evaluate HRQOL among patients diagnosed with DLBCL, using a novel nationally representative population-based dataset, and describe patterns in HRQOL by time from diagnosis.


The National Cancer Institute sponsored Surveillance, Epidemiology, and End Results-Medical Health Outcomes Survey (SEER-MHOS) linkage database is a research resource that allows for evaluation of HRQOL in cancer patients and survivors. The MHOS was administered annually to cohorts of patients randomly selected from Medicare Advantage health plans; each cohort was surveyed at baseline and 2 years later in follow-up. The MHOS measured HRQOL using the SF-36, an instrument with established reliability and validity for patients with cancer. The questions of the SF-36 capture data on 8 dimensions of general health that are grouped into physical (PCS) and mental (MCS) summary scores, capturing physical function and emotional well-being, respectively. Patients included in 6 cohorts (baseline 1998–2003; last follow-up on cohort 6 in 2005) were linked to the SEER database. For this cross-sectional analysis, we selected all patients age ≥65 that had a diagnosis of DLBCL (ICD-O-3 8680, 8684); HRQOL responses from the first available survey after DLBCL were used. SF-36 PCS and MCS (both scales range from 0–100) and poor self-rated health (self –reported fair or poor general health compared to other people your age) were compared among DLBCL patients by time from diagnosis to survey: 0–1, 1–3, and 3–5 years. HRQOL was also compared to patients that had MHOS data >1 year prior to a DLBCL diagnosis. Differences in PCS and MCS median scores were tested using the Wilcoxon rank sum test, and differences in the proportions of participants reporting fair or poor self-rated health were tested using the χ2statistic.


Median age and range for patients surveyed before, 0–1 (n=62), 1–3 (n=76), or 3–5 (n=31) years after their DLBCL diagnosis was similar (medians 73, 75, 74, and 78 years old, respectively; p=0.37). Date ranges for DLBCL diagnoses are as follows: 1997–2005, 1995–2004, and 1993–1998 for those surveyed 0–1, 1–3, and 3–5 year after diagnosis. HRQOL and self-rated health were low among the patients surveyed prior to their DLBCL diagnosis (n=296; PCS median=45.0, MCS median=56.3, poor self-rated health: 22.3%). In comparison, patients surveyed 0–1 year after DBCL diagnosis have even worse HRQOL scores (PCS median=33.6, MCS median=40.8, poor self-rated health: 51.6%; p=<0.0001 for all three comparisons). While older DLBCL patients surveyed 3–5 years from diagnosis have better quality of life (PCS median=36.0, MCS median=53.3, poor self-rated health fair: 29.0%) in comparison to the participants surveyed 0–1 year after diagnosis, PCS is still significantly different from the similarly aged group that was surveyed before their DLBCL (p=0.021).


In this SEER-MHOS population of older patients with a history of DLBCL, HRQOL is surprisingly low. Older patients prior to DLBCL had low scores on both the physical and mental components of the SF-36, indicating vulnerability prior to diagnosis and treatment. HRQOL and self-rated health was statistically and clinically worse among DLBCL patients less than a year out from diagnosis, possibly due to therapy. Notably, HRQOL remains low, compared to pre-diagnosis, in patients surveyed 3–5 years out from diagnosis, particularly in the physical domains. This study provides a benchmark for HRQOL among older patients with DLBCL, an NHL subtype in which aggressive treatment with curative intent is a standard approach, and further research that evaluates HRQOL prospectively in vulnerable older patients receiving treatment for DLBCL is critical.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.