The Role of 24‐Month Progression‐Free Survival (PFS24) in the Long‐Term Evaluation of Patients With Diffuse Large B Cell Lymphoma (DLBCL): A Real‐World Single Center Study

Diffuse large B cell lymphoma (DLBCL) is the most common aggressive B-cell lymphoma and is currently managed with chemoimmunotherapy as front-line treatment [1]. Approximately 60% of patients are cured, while relapses or refractory disease are typically observed within the first 2 years after diagnosis [2]. Large-scale studies have shown that patients with DLBCL who achieve progression-free survival 24 months after diagnosis (PFS24) have outstanding long-term outcomes, with an overall survival (OS) equivalent to that of the age- and sex-matched general population [3, 4]. To confirm previous reports, we examined long-term outcomes of a real-world Greek cohort of DLBCL patients, aiming additionally to determine baseline factors that could predict achievement of PFS24. Our original cohort included 419 consecutive patients, who were diagnosed with DLBCL from 2003 to 2022 and were treated with the combination of rituximab—cyclophosphamide, anthracycline, vincristine and prednisolone (R-CHOP) as first-line therapy. Of these, 258 patients were included in the analysis, including 102 who had an event within 2 years of follow-up (non-PFS24) and 156 who did not (PFS24 patients). The remaining 161 patients were excluded from the analysis, as they did not have an event and had not completed 2 years of follow-up. An event was defined as refractory disease upon first reassessment, progressive or relapsed disease or death. In addition, factors that could have a predictive role for the achievement of PFS24 were studied, including the International Prognostic Index (IPI) [5], the histological type of the cell of origin (COO) according to the Hans Algorithm [6] and the double expression (DE) status (bcl-2 > 50% immunohistochemical positivity and c-myc > 40% immunohistochemical positivity). The median age of the analyzed cohort was 67 years (range 5%–95%: 37–84), while the median age of the patients enrolled in the clinical trials of SEAL collaboration was 62 years [3]. Patients were 41.8% male and 54.2% had advanced disease (stage III–IV). IPI was available in 256 patients, with 71.1% having an IPI ≥ 2. The vast majority of the patients were diagnosed with DLBCL non otherwise specified (NOS) (95.3%). COO was available in 216 patients, in 37% of whom it was germinal center B cell (GCB). The baseline patient characteristics are depicted in Table 1. After a median follow-up of 3.9 years (range 5%–95%: 0.2–13.6), 192 (74.4%) patients achieved complete remission (CR). The percentage of alive patients was 62.3% (95% CI: 55.3%–68.2%). Importantly, during this period, 94.5% (87.6–97.6) of the patients who achieved PFS24 were alive, whereas only 10.9% (0.04–0.21) of the patients who did not achieve PFS24 (nonPFS24) were alive (Figure 1A). (A) Kaplan–Meier curve showing the overall survival of patients based on PFS24 status. (B) Percentage of patients achieving PFS24 per different groups. Capped bars denote confidence intervals. Asterisks denote p < 0.05. A total of 91 deaths were recorded, 85 in non-PFS24 patients and 6 in PFS24. Causes of death in PFS24 patients included disease relapse at 3 years after diagnosis in two patients, secondary acute myeloid leukemia (AML) in one patient, lung cancer in one patient and other causes in two patients. There was one patient who relapsed at 7 years of observation and achieved second remission after new treatment, in contrast to a similar study in a Dutch real-world population [7]. For non-PFS24 patients, the main cause of death was disease progression or relapse (61.2%), followed by sepsis (12.9%), pneumonia (7.1%), and gastrointestinal bleeding (1.2%), which mainly related to toxicity after first-line therapy or salvage therapy. Data on cause of death were not available for 17.6% of patients. Concerning factors that can affect PFS24, patients with IPI < 2 achieved PFS24 at a higher rate compared with patients with IPI ≥ 2 (85.1% [CI: 75–92.3] versus 51.1% [CI: 43.6–58.7] respectively p < 0.001). Additionally, patients with GCB achieved PFS24 at a higher rate compared with patients with non-GCB (68.8% [CI: 57.4–78.7] versus 55.2% [CI: 46.4–63.7] respectively p = 0.049). On the other hand, we found no effect of DE status on PFS24 achievement (42.9% [CI: 21.8–66] in double expressors versus 62.8% [CI: 52.2–72.5] in non-double expressors; p = 0.093). There was no difference regarding achievement of PFS24 according to COO in patients with IPI ≥ 2 (53.8% [CI: 39.5–67.8] in GCB versus 50% [CI: 39.5–67.8] in non-GCB respectively p = 0.73). However, a significant difference was observed regarding achievement of PFS24 according to COO in patients with IPI < 2 (96.4% [CI: 81.7–99.9] in GCB versus 72.7% in non-GCB [54.5–86.7] respectively p = 0.016). Results are presented in detail in Figure 1B. As previously observed in DLBCL patients included in clinical trials, OS was significantly better in DLBCL patients who achieved PFS24, which suggests its usefulness as a tool for the clinical evaluation and prognosis of patients after the end of frontline treatment with anthracycline-containing regimens. COO determined by immunohistochemistry using the Hans algorithm may have a prognostic value in DLBCL patients treated with R-CHOP although reports from the current literature are conflicting [8, 9]. IPI as well as COO could predict achievement of PFS24 in our patient cohort. Specifically in patients with low-risk lymphoma based on IPI, histologic subtype according to Hans' algorithm may have an important predictive role for PFS24 achievement. In our cohort, non-GCB patients were less likely to achieve PFS24 compared to GCB patients in the IPI < 2 group. The dismal prognosis in this subset of patients could justify the use of novel regimens in first-line treatment such as Polatuzumab Vedotin-R-CHP [10]. Overall, despite the relatively small number of patients in this cohort, our results may imply some novel predictive tools in DLBCL patients that deserve further investigation. The authors declare no conflicts of interest. The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1002/hon.70034. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.