Patterns of Care in a Random Population-Based Sample of Patients Diagnosed with Non-Hodgkin’s Lymphoma.

The advent of therapeutic monoclonal antibodies has enhanced the efficacy of NHL treatment. In recent years, these immuno-therapies have been increasingly used in therapy. We conducted a population-based study of NHL treatment practices in the US using a stratified random sample of patients diagnosed in 1999 with histologically confirmed NHL (n=939) residing in the geographic areas covered by the Surveillance, Epidemiology and End Results program. Blacks and Hispanics were over-sampled to obtain more stable estimates. Patients were followed for vital status through Dec 2001. We performed separate logistic regression analyses to study the potential factors associated with the likelihood of receiving chemotherapy, radiation therapy and the monoclonal antibody, Rituximab. Cox Proportional Hazards regression model was used to study the risk factors associated with survival time. We grouped histological subtypes into five broad categories: B-cell aggressive, B-cell indolent, T-cell generic, cutaneous T-cell, and mantle cell lymphomas. The majority of patients presented with B-cell aggressive or B-cell indolent lymphomas (n=828). Approximately 20% of patients received no therapy. Over 60% of patients received chemotherapy, either alone or in combination. 12% of patients received Rituximab and it was most frequently administered to patients in combination with chemotherapy, especially for patients with B-cell aggressive, B-cell indolent and T-cell generic lymphomas. Only 3% of patients participated in clinical trials. Age and gender were associated with the receipt of chemotherapy: people aged over 75 years, and males were less likely to have received chemotherapy (P=0.01). There were no significant associations between the likelihood of receiving Rituximab and the demographic and clinical factors analyzed. However, our results suggested that African-Americans and people aged over 75 years were less likely to have received immunotherapy. Twenty-four percent of patients received radiation with or without another therapy. When compared to patients with no symptoms at presentation, patients who presented with B-symptoms at diagnosis or those whose B-symptoms were unknown were less likely to have received radiation therapy (OR=0.32 and 0.47 respectively, P=0.0002). Approximately 50% of patients had died by the end of maximum the 3-year follow-up period. Both cause-specific and all-cause mortality was significantly associated with patient age, race/ethnicity, gender, marital status and co-morbid conditions, as well as histological subgroup. Hispanic and Black patients had higher risk of death from both NHL and all-cause (P<0.01) than their non-Hispanic white counterparts. Patients > 75 years, male patients, unmarried patients, or patients with B-symptoms had higher risk of death from either NHL or all-cause (p<0.01). This paper is the first population-based study examining the receipt of therapy for many histological subtypes of NHL. Future work will examine the impact of treatment on survival.