Effectiveness and toxicity of cyclophosphamide, etoposide, and prednisone as an outpatient oral chemotherapy regimen in elderly and frail patients with non-Hodgkin lymphoma

Abstract INTRODUCTION The management of non-Hodgkin lymphoma (NHL) in elderly or frail patients not eligible for intensive chemotherapy remains a significant clinical challenge due to increased comorbidities, higher risk of treatment-related toxicity, and the logistical burden of frequent hospital visits required for intravenous chemotherapy administration. At our institution, an oral chemotherapy regimen based on cyclophosphamide, etoposide, and prednisone (CEP) is employed as an alternative therapeutic approach. The objective of this study aims to evaluate the efficacy and safety profile of the CEP regimen in this vulnerable patient population. METHODS This retrospective, single-center observational study involved all patients with NHL aged ≥70 years who received treatment with the CEP regimen between 2018 and 2024. The regimen consists of fixed oral doses administered in 21-day cycles: cyclophosphamide 150 mg/day for 5 days, etoposide 100 mg/day for 3 days, and prednisone 100 mg/day for 5 days. Rituximab, at the standard dose, was incorporated into the treatment protocol in selected patients, as determined by physician discretion. RESULTS The study included 61 patients, with a median age of 82 years (range 70–92), and 34 (55.7%) were women. The most frequent lymphoma subtype was diffuse large B-cell lymphoma (DLBCL) in 29 patients (47.5%), followed by follicular lymphoma (FL) in 16 patients (26.2%), T-cell non-Hodgkin lymphoma in 7 patients (11.5%), marginal zone lymphoma in 5 patients (6.6%), and mantle cell lymphoma in 4 patients (6.6%). Only 9 patients (14.7%) presented with limited-stage disease (stage I or II). CEP was used as first-line therapy in 34 patients (55.7%) and the median number of cycles administered was 5 (range 1–15). Rituximab was included in the treatment regimen of 33 (61.1%) patients with B-cell lymphoma. At the initiation of CEP treatment, 25 patients (41%) had an ECOG performance status ≥2, all patients had a Charlson Comorbidity Index ≥5, and 55 patients (90.2%) scored ≤14 on the G8 scale. The overall response rate (ORR) was 55.7%, with complete responses (CR) observed in 20 patients (32.8%). Early progression occurred in 16 patients (26.2%), leading to premature treatment discontinuation. The median follow-up was 38.6 months. Median overall survival (OS) and progression-free survival (PFS) were 19.0 and 9.6 months, respectively. Among the most frequent lymphoma subtypes in our cohort, DLBCL showed an ORR of 58.6% and a CR rate of 44.8%, with a median OS of 19.1 months and PFS of 8.1 months. In FL, the ORR was 62.5% and the CR rate was 25%, with median OS and PFS of 40.2 and 38.4 months, respectively. Factors that significantly influenced ORR, OS, and PFS were: having an ECOG performance status <2 at the start of treatment (ORR 70.6% vs. 29.4%, median OS 25.8 vs. 6.0 months, median PFS 22.6 vs. 5.0 months); receiving the CEP regimen as first-line therapy (ORR 70.6% vs. 29.4%, OS 35.5 vs. 6.4 months, PFS 19.1 vs. 3.5 months); and the addition of rituximab in patients with B-cell lymphomas (ORR 80.6% vs. 19.4%, OS 35.5 vs. 5.8 months, PFS 22.6 vs. 3.4 months). A higher G8 score (analyzed as a continuous variable) was significantly associated with increased ORR(OR 0.79, 95% CI 0.63–0.98),and prolonged OS(HR 0.8, 95% CI 0.71–0.93) and PFS(HR 0.85, 95% CI 0.74–0.97). At least one treatment-related adverse event was reported in 41 patients (67.2%), with grade ≥3 infections (26.2%) and hematologic toxicities (16.4%) being the most frequent. Febrile neutropenia occurred in only 3 patients (4.9%), and rituximab-associated pneumonitis was documented in 2 patients (3.3%). Notably, 40 patients (65.6%) required neither dose adjustments nor delays between treatment cycles, and treatment was discontinued due to toxicity in only 12 patients (19.7%). During follow-up, 44 patients (72.1%) died, with disease progression being the primary cause (35 cases, 57.4%). Three patients (4.9%) died due to treatment-related respiratory infections. CONCLUSIONS The oral CEP regimen constitutes a well-tolerated outpatient treatment with an acceptable toxicity profile, representing a viable therapeutic option for elderly and frail patients with NHL. In our cohort, it achieved an ORR of 55.7% and a median OS of 19 months. The regimen showed enhanced efficacy when administered as first-line therapy and in combination with rituximab for B-cell lymphomas, both of which were associated with a significant improvement in overall survival.