Impact of Progression-Free Survival at 24 Months on Subsequent Survival in Patients with Diffuse Large B-Cell Lymphoma Treated with R-CHOP Therapy: A Supplementary Analysis of JCOG0601

Abstract

Introduction The impact of progression-free survival status at 24 months (PFS24) or event-free survival status at 24 months (EFS24) on subsequent survival has been evaluated in patients with various lymphoma subtypes. Patients with newly diagnosed diffuse large B-cell lymphoma (DLBCL) who achieve PFS24 or EFS24 generally have excellent outcomes. It has been observed that the overall survival (OS) after achieving EFS24 was not significantly different from that of the general population (M.J. Maurer. J Clin Oncol, 2014). However, the OS after achieving PFS24 was significantly worse than that of the general population (M.J. Maurer. Ann Oncol, 2018). Consequently, the impact of achieving PFS24 for DLBCL patients remains controversial. In this supplementary analysis, we evaluated the impact of achieving PFS24 on the subsequent survival of untreated DLBCL patients by using data from JCOG0601, a randomized phase2/3 study assessing the schedule of rituximab administration in combination with tri-weekly CHOP (Ohmachi K. Blood Adv, 2021). Methods Among 422 patients enrolled between Dec 2007 and Dec 2014 in JCOG0601, 409 patients were eligible for this analysis (Data cutoff date: Dec 19, 2017). PFS24 was defined as being alive without progression or relapse for 24 months from randomization. The protocol treatment was initiated within 7 days after randomization. OS from PFS24 was defined as the time from achieving PFS24 or the date of progression or relapse to death from any cause. We compared OS from PFS24 with that of age-, sex-, and calendar period-matched Japanese general population using standardized mortality ratios (SMRs). Similarly, PFS12 and PFS60, as well as OS from PFS12 and PFS60, were defined in the same manner as PFS24 and OS from PFS24. The log-rank P-values for OS were calculated. Results The baseline characteristics of the 409 patients were as follows: a median age of 62 years (range, 20-79), with males accounting for 56% (n=227) of the patients. At diagnosis, 46% (n=188) of the patients were Ann Arbor stage III or IV. The majority of patients (82%; n=335) were classified as international prognostic index low or low-intermediate risk. Based on the Hans algorithm for cell-of-origin, 51% (n=210) of the patients were categorized as non-germinal center B-cell type (non-GCB). At a median follow-up of 5.3 years among all patients, a total of 334 patients (82%) achieved PFS24, while 66 patients (16%) failed to achieve PFS24. Five patients died without progression within 24 months, and four patients were lost to follow-up within 24 months. Multivariable analysis revealed that risk factors for failing to achieve PFS24 included serum LDH levels higher than the upper normal limit and two or more extranodal lesions. Patients who achieved PFS24 had a significantly better OS than those who failed to achieve PFS24 (median OS, not reached vs. 1.3 years; P < 0.001). Similar results were observed when using PFS12 and PFS60, regardless of the cell-of-origin. Among the patients who failed to achieve PFS24, non-GCB type patients more frequently had relapsed or refractory diseases in extranodal sites (72%: n=26/36) compared to GCB type patients (43%; n=6/14). The OS after achieving PFS24 or PFS60 was not remarkably different from that of the general population (PFS24: SMR 1.29, 95% confidence interval [CI] 0.72-2.12, P = 0.39; PFS60: SMR 1.43, 95% CI 0.47-3.33, P = 0.55). However, the OS after achieving PFS12 was significantly worse than that of the general population (SMR 2.30, 95%CI 1.59-3.22, P < 0.001). The primary cause of death for patients achieving PFS12 was DLBCL with the cumulative incidence of more than 5% at 5 years, while the incidence of death due to DLBCL was less than 5% for those achieving PFS24. The primary cause of death for the patients achieving PFS24 was other diseases except for DLBCL and treatment-related toxicity, including secondary malignancies (n=4/7) and pneumonia (n=3/7). Conclusion Newly diagnosed DLBCL patients treated with R-CHOP who achieved PFS24 exhibited an excellent subsequent outcome, which did not differ from those of age-, sex-, calendar-period matched individuals in the general population. Our findings suggest that PFS24 achievement could serve as a surrogate endpoint for OS in DLBCL patients. Further research is warranted to establish the utility of PFS24 achievement as a reliable milestone in clinical practice.