Immunotherapy may be preferable to chemotherapy for patients with <scp>POD24</scp> follicular lymphoma in the second‐line setting

Follicular lymphoma (FL) is characterized by a relapsing course and eventual refractoriness to therapy. Early progression of FL after first-line treatment is associated with poor outcomes, especially for patients who progress within 24 months of initial therapy (POD24).1, 2 A considerable fraction of these early events correspond to histological transformation (HT), as demonstrated in the PRIMA study and other contemporary series, where most HT episodes occurred within the first years after induction therapy.3-5 However, some POD24 patients present a prolonged survival after first relapse, especially those with a low/intermediate Follicular Lymphoma International Prognostic Index (FLIPI) score and without HT at first relapse.6 Published reports identified male sex, poor performance status (PS), high FLIPI score at diagnosis and elevated β2 microglobulin levels as predictive factors of POD24. Nevertheless, tools for assessing POD24 risk at diagnosis remain suboptimal and are not currently implemented in clinical practice.7 Most studies validating the prognostic impact of POD24 have focused on patients who received chemotherapy (CT) as first-line treatment.8, 9 In chemo-free settings, data are more limited but consistent. In the RELEVANCE trial, POD24 remained associated with an adverse prognosis irrespective of whether patients received lenalidomide-rituximab or chemoimmunotherapy.10 Similar findings have been reported in studies evaluating other first-line immunotherapy (IT)-based approaches.11, 12 The optimal second-line treatment for these early relapsing patients is not well established, leading to heterogeneous management across centres. Several regimens have demonstrated efficacy in relapsed/refractory FL, including obinutuzumab–bendamustine (GADOLIN)13 and lenalidomide–rituximab (AUGMENT),14 while CD20xCD3 bispecific antibodies are reshaping treatment strategies.15 Currently, it remains unclear whether novel IT approaches improve patient outcomes compared to CT regimens in second-line POD24 patients. The aim of our study was to analyse treatment patterns in this setting, comparing outcomes of POD24 patients who received CT versus IT-only strategies. We conducted a retrospective analysis of patients diagnosed with FL grade 1-3a who met the POD24 definition, treated with upfront (CT) in 16 GELTAMO (Grupo Español de Linfomas y Trasplante de Médula Ósea) centres from January 2015 to April 2024. Patients who received single-agent monoclonal antibodies or radiotherapy as first-line treatment were excluded from the analysis. Patients were categorized into two groups according to the second-line regimen: the CT group, which included those who received CT as part of the second-line regimen; and the IT group, which included those who received monoclonal antibodies with lenalidomide and/or CD20xCD3 bispecific antibodies, without a CT backbone. We also collected data on patients who presented HT to large B-cell lymphoma (LBCL) after second-line treatment. This study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Clinical Research Ethics Committee of the Vall d'Hebron University Hospital (EOM[AG]025/2024[6261]). Outcomes of interest were overall and complete response rates (ORR/CRR), duration of response (DOR), progression-free survival (PFS) and overall survival (OS) using the Kaplan–Meier method and the Cox model. An inverse probability of treatment weighting (IPTW) ATT analysis was performed to adjust for potential imbalances (selection bias) between CT and IT cohorts. Variables included in the IPTW adjustment were: age, maintenance therapy after first-line, stage at first relapse, FLIPI at first relapse, autologous stem cell transplantation (ASCT), B symptoms at first relapse and best response to first-line treatment (Table S1). Among 111 POD24 FL patients, 86 received CT and 25 received IT as second-line treatment. Baseline characteristics were analysed according to the first-line and second-line treatment they received (Table 1). For the full cohort, the median age for the overall cohort was 57 years, with no differences according to treatment approach (60 vs. 53 years, p = 0.14). Fifty-four patients (49%) were female and 103 (96%) presented an advanced stage disease. Most of the patients in the CT and IT groups had an intermediate- and high-risk FLIPI score at diagnosis (93% vs. 88%, p = 0.6) and at time of first relapse (67% vs. 88%, p = 0.5) respectively. Distribution of first-line treatment was well balanced between both groups. CHOP-based regimens were administered to 92% of patients in the CT group and 88% of patients in the IT group. Also, bendamustine in combination with rituximab was administered to 8% and 12% of the CT and IT groups respectively. No patients received the combination of bendamustine and obinutuzumab as first-line treatment. In terms of response to frontline therapy in the full cohort, 29 patients (26%) attained a complete remission (CR), 51 patients (46%) a partial response (PR) and 31 patients (28%) experienced a progressive disease (PD) as their best response from the overall cohort. There was a higher rate of PD as best response after first-line treatment in the CT group than in the IT group (34% vs. 8%, p = 0.026). Maintenance treatment was administered to a total of 75 patients (68%), with lower use in the CT versus IT groups (50 patients [58%] vs. 22 patients [88%], p = 0.008). Survival outcomes in the subgroup of patients who received first-line maintenance are shown in Figure S2. Second-line treatment in the CT group included bendamustine-based regimens (n = 37, 43%), R-ESHAP (n = 33, 38%), gemcitabine-based regimens (n = 4, 5%) and other regimens (n = 12, 14%). Comparative survival outcomes for the main CT subgroups (bendamustine-based and R-ESHAP) versus IT are presented in Figure S3. The IT cohort comprised patients treated with lenalidomide in combination with monoclonal antibodies (n = 9, 36%), bispecific antibodies (n = 15, 60%) and idelalisib (n = 1, 4%). For the full set of patients, best response included CR in 54 patients (49%), PR in 20 (18%) and PD in 37 (33%). There were no significant differences in terms of ORR (65% vs. 72%, p = 0.52) or CRR (45% vs. 60%, p = 0.2) between the CT and IT groups. ASCT consolidation was performed in 27 patients (25%) (29% for the CT and 8% for the IT, p = 0.034). In terms of HT, it was reported in 15 patients (14%), all of whom had received CT as second-line treatment. HT was evidenced after a median of 41 months from initial diagnosis (range, 8–61 months) and after a median of 25 months from the beginning of the second-line treatment (range, 1–48 months). Efficacy of second-line treatment in the POD24 population regarding both cohorts was analysed (Figure S1). With a median follow-up from diagnosis of 46.1 months for the overall cohort, 58.3 months for the CT group and 27.9 months for the IT group, median PFS for CT patients was 12.3 months (95% confidence interval [CI]: 7.8–19.2) versus not reached for IT patients (hazard ratio [HR]: 0.42, 95% CI: 0.21–0.84; p = 0.014). These results were confirmed in the IPTW analysis (Figure 1A). The median DOR for patients receiving CT was 16.5 months (95% CI: 9.9–24.8), whereas it was not reached for patients receiving IT (HR: 0.22, 95% CI: 0.07–0.73; p = 0.013); these results were also confirmed in the IPTW analysis (Figure 1B). The median OS for CT and IT patients was not reached, but there was a trend towards prolonged survival for the IT group (HR: 0.16, 95% CI: 0.02–1.17; p = 0.07). However, these results were not confirmed in the IPTW analysis (Figure 1C). In our dataset, IT regimens as second-line treatment in POD24 FL patients were associated with an improved outcome compared to standard CT. These findings could justify our hypothesis that POD24 patients after frontline CT have a chemorefractory disease and may benefit from bispecific antibodies or lenalidomide combinations. Direct comparisons between CT and IT in POD24 are limited. Our analysis provides relevant insights into optimal second-line management. For patients receiving CT as second-line, ASCT consolidation remains an option for selected cases achieving a response. However, its use has declined in recent years due to the emergence of novel strategies. Regarding HT, it was confirmed in 14% of the patients included in our full cohort; all of them were treated with CT in second line before HT. Tools to identify at diagnosis individual patients who will eventually meet POD24 criteria or will eventually experience HT are not well defined. Limitations include retrospective design, the limited sample size in the IT cohort and the heterogeneity in treatment regimens. Additionally, pathology was not centrally reviewed and a higher proportion of patients in the CT group presented HT, suggesting a more refractory disease course, which may underlie worse outcomes independently of second-line therapy. First-line therapy was also heterogeneous, with a predominance of CHOP-based regimens in several centres; although POD24 is largely driven by adverse biological features rather than the specific frontline regimen, this variability could have contributed to differences in post-relapse outcomes. Furthermore, because participating centres contributed only eligible POD24 cases rather than all consecutive FL diagnoses, the incidence of POD24 could not be estimated and no reliable denominator of the underlying FL population was available. In summary, second-line treatment approaches for POD24 FL patients are heterogeneous across centres. In our study, a prolonged DOR and PFS were observed in the group of IT, compared to CT. More follow-up is needed to confirm the observed OS trend in favour of the IT cohort. Novel agents such as lenalidomide and bispecific antibodies, alone or in combination, seem to improve the outcomes of this high-risk patient population. Future research must focus on integrating these novel treatment approaches into standard algorithms and identifying optimal sequencing to improve the balance of efficacy and safety. Serna Ángel, Navarro Víctor and Abrisqueta Pau conceived the idea and designed the study. Serna Ángel, Bosch Marc, López-Pereira Patricia, Quintela David, Sancho Juan-Manuel, Sáez Adolfo, Jiménez-Ubieto Ana, Caballero Juan Carlos, Córdoba Raúl, Cerdà María, De la Cruz Beatriz, Romero Samuel, Ferrero Ainara, García Tomás, Villalobos Teresa, Rivero Andrea, Vázquez Sofía, Villafuerte Paola, Herrero Carmen, Canales Miguel, Martín García-Sancho Alejandro and Marco Javier supplied the study material or patients. Serna Ángel, Bosch Marc, López-Pereira Patricia, Quintela David, Sáez Adolfo, Caballero Juan Carlos, Canales Miguel, De la Cruz Beatriz, Romero Samuel, Villalobos Teresa, Ferrero Ainara, Vázquez Sofía, Villafuerte Paola, Herrero Carmen, Alcoceba Miguel, Martín García-Sancho Alejandro and Marco Javier collected and assembled the data. Serna Ángel and Navarro Víctor performed the statistical analysis. Serna Ángel, Navarro Víctor, Iacoboni Gloria, Bosch Francesc and Abrisqueta Pau analysed and interpreted the data. Serna Ángel, Navarro Víctor and Abrisqueta Pau wrote the manuscript. All authors reviewed and approved the manuscript. We thank all participating GELTAMO centres and their clinical teams for their collaboration in data collection and patient management. We are also grateful to the statistics unit at Vall d'Hebron Institute of Oncology (VHIO) for methodological support. The authors acknowledge the contribution of all patients and families, whose participation made this study possible. This research received no specific grant from any funding agency, commercial or not-for-profit sectors. Á. Serna: Honoraria from Johnnson&Johnson, Incyte, AbbVie, Behring, Lilly, Roche and Astrazeneca. M. Bosch: No conflicts of interest. V. Navarro: No conflicts of interest. P. López-Pereira: Honoraria from Johnson&Johnson, Roche, Beigene and Takeda and advisory for Roche. D. Quintela: No conflicts of interest. J-M Sancho: Honoraria from Roche, Gilead, Johnson&Johnson, Celgene, Servier, Novartis, Mundipharma, Sanofi and Kern-Pharma and advisory for Roche, Janssen, Gilead, Celgene, Bristol Myers and Celltrion. A. Sáez: No conflicts of interest. A. Jiménez Ubieto: Consultancy from Regeneron, Lilly, Genmab, Kite/Gilead, Roche, AbbVie, Speakers bureau from Sandoz, Incyte, Kite/Gilead, Roche, AbbVie. J-C Caballero: No conflicts of interest. R. Córdoba: Research Funding from Pfizer; Advisory from Kyowa-Kirin, Roche, AbbVie, ADC Therapeutics, Bristol-Myers Squibb, Incyte, Takeda, Johnson&Johnson and Kite; Speakers Bureau from Roche, AbbVie, Bristol-Myers Squibb, Takeda, Janssen and Kite. M. Cerdà: No conflicts of interest. B. De la Cruz: Advisory role from Kyowa-Kirin and Roche. Honoraria from Kyowa-Kirin, Johnson&Johnson, Sandoz and AstraZeneca. S. Romero: Honoraria from AstraZeneca, Incyte, Johnson&Johnson, Kite, Kyowa-Kirin; Advisory from Roche and Kiowa-Kirin. A. Ferrero: No conflicts of interest. T. García: Consultant or advisory role from Johnson&Johnson, Roche, Gilead and Celgene; Research funding from Johnson&Johnson and AbbVie. T. Villalobos: No conflicts of interest. A. Rivero: No conflicts of interest. M-S Vázquez: No conflicts of interest. P. Villafuerte: No conflicts of interest. C. Herrero: No conflicts of interest. M Canales: Consultancy from Karyopharm, Johnson&Johnson, Incyte, BMS, BeOne, Kyowa, Novartis, Sanofi, Takeda, Roche; Speakers bureau from Johnson&Johnson, Kite/Gilead, Kyowa, Novartis, Roche, Takeda, Sandoz, Amgen. M. Alcoceba: No conflicts of interest. A. Martín García-Sancho: Consultancy from AbbVie, AstraZeneca, BeOne, BMS, Genmab, Gilead/Kite, GSK, Ideogen, Johnson&Johnson, Lilly, Miltenyi, MSD, Regeneron, Roche, Sobi. Honoraria from AbbVie, AstraZeneca, BeOne, BMS, Genmab, Gilead/Kite, GSK, Ideogen, Johnson&Johnson, Kyowa Kirin, Lilly, MSD, Regeneron, Roche, Sobi, Takeda; Research suport (GELTAMO) from Gilead/Kite, Roche. J. Marco: No conflicts of interest. C. García: Honoraria from Roche and AbbVie. A. Marín-Niebla: Consultant or advisory role from Lilly, Kite, Kyowa-Kirin, Takeda, AstraZeneca, Roche and Johnson&Johnson; Honoraria from Lilly, Kite, Takeda, Janssen. G. Iacoboni: Honoraria from Celgene/BMS, Autolus, Novartis, Roche, AbbVie, Johnson&Johnson and Gilead/Kite, AstraZeneca, Miltenyi, Regeneron. F. Bosch: Consultant or advisory role from Novartis, Celgene/BMS, Roche, AstraZeneca, Takeda, Mundipharma, Gilead, AbbVie, Lilly and Beigene; Honoraria from Novartis, Celgene/BMS, Roche, AstraZeneca, Takeda, Karyospharm, Mundipharma, Gilead, AbbVie, Lilly and Beigene. P. Abrisqueta: Consultant or advisory role from Roche, AbbVie, BMS, AstraZeneca, Johnson&Johnson and Incyte; Honoraria from Roche, AbbVie, BMS, AstraZeneca, Johnson&Johnson and Sandoz. This study was conducted in accordance with the Declaration of Helsinki and approved by the Clinical Research Ethics Committee of Vall d'Hebron University Hospital (EOM[AG]025/2024[6261]). Use of patient data for research was authorized under institutional informed consent procedures approved by the Clinical Research Ethics Committee of Vall d'Hebron University Hospital. The data that support the findings of this study are available from the corresponding author upon reasonable request. Data S1. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.