Efficacy and Safety of CHOP-14 in Elderly Patients with Very Poor Risk Diffuse Large B-Cell Lymphoma.

The outcome of advanced diffuse large cell B-cell lymphoma (DLBCL) has been improved by the “dose-densified” biweekly CHOP-14 regimen. The German NHL-B2 trial has demonstrated favourable efficacy and safety in elderly patients with performance status ≤ 3 and normal organ function. However, a considerable proportion of patients are not eligible for clinical trials due to high age, poor performance, concomitant disease and/or organ dysfunction. The efficacy and safety of CHOP-14 is unknown in this group of very poor risk patients. CHOP-14 was introduced as standard treatment for advanced DLBCL (CSII-IV and/or LDH above UNV and/or bulky tumor ≥ 7.5 cm) at our institution in 2002. Seventy consecutive patients with DLBCL have been treated with 6-8 series CHOP-14 in the period March 2002 to December 2003. Patients with residual disease following chemotherapy were subsequently treated with involved field radiation. In order to estimate the efficacy and safety of CHOP-14 in very poor risk patients, we divided this population into two cohorts; A: standard risk: pts. aged 60–75 years with PS ≤ 3 or pts. aged < 60 years regardless of PS and B: very poor risk: pts. aged 60–75 years with PS = 4 or pts. aged > 75 years regardless of PS.

Patient characteristics

The response rates in the two cohorts (A vs. B) were CR/Cru: 82% vs. 62% (p=0.06); PR: 2% vs. 8% (p=0.55); NC/PD: 11% vs. 8% (ns). Two year EFS was 66% vs. 37% (p=0.027) and the two year OS was 71% vs. 57% (p=0.04). Concerning toxicity 30/44 (68%) vs. 23/26 (88%) (p=0.08) required hospitalisation for one or more of the following reasons: 50% vs. 65% (p=0.21) due to infection; 9% vs. 15% (p=0.42) due to Pneumocystis Carinii pneumonia; 27% vs. 31% (p=0.75) due to reduced PS mainly caused by malnutrition. The median total number of days in hospital were 8 (1–162) vs. 39 (1–228) days (p=0.002). The therapy-associated deaths without progression were 2/44 (5%) in cohort A (1 cardiac arrest, 1 ileus) compared to 3/26 (12%) in cohort B (1 cardiac arrest, 1 PCP, 1 unknown) (ns). In 3 additional patients in cohort B, treatment was stopped early due to severe infection, resulting in progression and subsequently death. In general dose erosion was minimal, except for vincristin that was reduced due to neuropathy in 20/44 (45%) vs. 10/26 (38%) (p=0.57) of the patients after a median of 5 vs. 4 cycles. The median delay in treatment schedule (schedule erosion) was 8 days (0–95 days) vs. 14 days (0–67 days) (p=0.06). Given the dismal prognosis of the patients in cohort B, the two-year survival rate of 57% is encouraging. However, the increased toxicity with infections and malnutrition, warrants for careful attention to this high risk group of patients, preferably within clinical trials focused on prevention and treatment of infections, improvement of the nutritional status and quality of life.