Search for a command to run...
Mantle cell lymphoma (MCL) is an aggressive and incurable disease.1, 2 Most patients are over 65 years old and exhibit heterogeneous fitness status alongside multiple comorbidities. Until now, conventional chemo-immunotherapy (CIT) followed by rituximab maintenance has represented the standard approach for elderly patients.3 Approximately 40% of MCL patients experience disease relapse or progression (POD) within 24 months from diagnosis and have been defined as early-POD.4 The MANTLE-FIRST study has divided patients' outcome depending on time to first relapse (early- vs. late-POD) after intensive upfront therapy including high-dose cytarabine.5, 6 However, the predictive significance of early-POD in elderly MCL patients following less intensive frontline treatment is not well established. Recent studies have addressed this topic, but only as pooled analysis from clinical trials or retrospective series including also young patients.7-9 In our retrospective real-life study, we evaluated the prognostic significance of early-POD specifically among relapsed/refractory (r/r) MCL patients ≥ 65 years old. We evaluated progression-free survival and overall survival, estimated from the start of salvage therapy (PFS-2, OS-2) in the whole cohort and according to different second-line treatments. We also evaluated PFS-2 and OS-2 in relation to a simplified geriatric assessment (sGA) tool, uniformly assessed in all participating Centers (Table S1).10 Patient selection/statistics/ethics are reported in the Supporting Information. We included 231 patients ≥65 years old with r/r disease after rituximab-based non-intensive induction regimens, who were treated at relapse with at least one cycle of systemic therapy (Bruton's tyrosine kinase inhibitor [BTKi] vs. CIT vs. other) between 2007 and 2021 in 21 centers belonging to the Fondazione Italiana Linfomi (FIL) (Table S2). Overall, 121 patients (53%) were defined as early-POD. Patients with older age at diagnosis, male sex, Eastern Cooperative Oncology Group performance status (ECOG PS) > 1, high lactate dehydrogenase and mantle cell lymphoma international prognostic index (MIPI), Ki67 > 30%, and nonclassical histology were more represented in the early-POD subgroup (Table S3). Frontline treatments included rituximab–bendamustine (RB = 102 patients, 45%), rituximab, bendamustine, and cytarabine (R-BAC = 67 patients, 29%), and rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP/R-CHOP-like = 60 patients, 26%). Patients treated with RB were older compared to R-BAC and R-CHOP, with a worse fitness profile. We showed more interruptions/dose reductions during R-BAC, with a higher incidence of hematological adverse events (AEs) (Table S4). Table S5 describes the impact of patients' characteristics/type of first-line treatment on time to POD, highlighting the relevance of age, ECOG PS > 1, and Ki67 > 30%. We then focused on the safety and effectiveness of different second lines. Patients' characteristics at relapse (n = 223) are reported in Table 1. Overall, 124 patients (56%) were treated with BTKi (122 ibrutinib, 1 zanubrutinib, and 1 acalabrutinib). Patients treated with BTKi were older than patients treated with salvage CIT (median age 77 vs. 75, P-value = 0.026, 40% ≥ 80 years) and unfit/frail in over 50% of cases. The remaining clinical features were equally distributed. BTKi showed less hematological toxicity, but slightly more extrahematological AEs. Cardiological toxicity was more frequent among the BTKi subgroup (four atrial fibrillation, two heart failure, and one acute myocardial ischemia). Among 90 patients who discontinued BTKi, the cause was progression in 61 (68%), intolerance in 9 (10%), and other reasons in 20 (22%) (n = 16 unknown, n = 3 secondary neoplasia, and n = 1 bridge to allogeneic transplantation). PFS-2 and OS-2 are reported in Figure S1. With a median follow-up of 4.6 years, median PFS-2 was 1.0 year (95% CI 0.76–1.3), and median OS-2 was 2.2 years (95% CI 1.6–2.6). Causes of death were 106 progressions (69%, n = 1 to central nervous system [CNS]), 12 infections (8%), 6 second cancer (4%), 4 heart failure (3%), 1 kidney failure, and 23 unknown (15%). PFS-2 and OS-2 according to different second-line treatments are shown in Figure 1A,B. Patients treated with BTKi had a median PFS-2 of 1.39 years, and 3-year PFS-2 (3y-PFS-2) of 24% versus a median PFS-2 of 0.98 years and 3y-PFS-2 of 11% for patients treated with CIT (hazard ratio [HR] 1.43, 95% CI 1.02–2.00, P-value = 0.037). Cox regression, adjusting second-line treatments by confounding factors such as MIPI and early-POD, is shown in Table S6, and confirmed the independent favorable impact of BTKi on PFS-2. In terms of OS-2, BTKi did not show a significant advantage (median OS-2 2.10 years and 3-year OS-2 [3y-OS-2] 36% vs. median OS-2 3.16 years and 3y-OS-2 53% for CIT; HR 0.82, 95% CI 0.57–1.20, P-value = 0.314). We then evaluated the role of time to POD on survival outcomes (Figure 1C,D). Among early-POD cases, median PFS-2 was 0.66 years versus 1.47 years for late-POD (HR 1.49, 95% CI 1.11–2.01, P-value = 0.008). Median OS-2 was 1.44 years for early-POD versus 3.17 years for late-POD cases (HR 1.55, 95% CI 1.12–2.15, P-value = 0.009). Well-known prognostic factors such as ECOG PS, MIPI, and CNS involvement confirmed their role at relapse (Tables S7 and S8). Unfit/frail patients showed a significantly worse outcome with median PFS-2 0.76 years (95% CI 0.41–1.16) versus 1.16 years (95% CI 0.92–1.95) for fit patients. 3y-PFS-2 was 11% (95% CI 5–20) in unfit/frail patients versus 32% (95% CI 22–42) in fit cases (P-value = 0.007). 3y-OS-2 was 26% (95% CI 17–36) compared to 47% (95% CI 35–57) in fit cases (P-value = 0.003) (Figure 1E,F). Finally, we considered the interaction between sGA and time to POD assessing the outcome of different subgroups in terms of 3y-PFS-2 and 3y-OS-2. Fit patients with late relapse (Fit-Late subgroup) had the best prognosis in terms of PFS-2 and OS-2 (34% and 62%, respectively). Among unfit/frail patients, 3y-PFS-2 and OS-2 were quite similar, regardless of time to POD (Table S9). At multivariable analysis, second-line treatment with CIT (compared to BTKi), unfit/frail status (both for early- and late-POD), high MIPI, and CNS disease at relapse were associated with worse PFS-2 (Table S10). In terms of OS-2, fit patients with early-POD and unfit/frail patients (both early- and late-POD), together with high MIPI and CNS relapse, confirmed their independent prognostic significance. Treatment with BTKi was associated in multivariable analysis with improved PFS-2 but not OS-2. Figure S2 shows predicted PFS-2 and OS-2 by multiple Cox regression, adjusted for sGA-POD, MIPI, and CNS disease at relapse. Only data about OS were available for 144 r/r patients after second-line therapy. Median OS-3 (overall survival after second relapse) was 6.4 months (95% CI 4.4–10.4). Thirty-five patients received BTKi as third line: only 6 were re-treated after prior BTKi, while 23 and 9 cases were previously treated with CIT or other schemes. As shown in Figure S3, patients receiving BTKi as third-line treatment after different prior therapies had a less unfavorable outcome (median OS-3 20 months). At the time of our study, chimeric antigen receptor T cells (CAR-T cells) and pirtobrutinib were not yet available in Italy for r/r MCL.11, 12 Moreover, most of our patients would not have been eligible for CAR-T due to unfitness. The present study described the adverse outcome of elderly MCL patients in the case of early-POD. To our knowledge, this is the first report describing the role of early-POD specifically in elderly MCL patients and the use of a simple and objective tool for fitness assessment in MCL, as part of daily clinical practice. The fitness status emerged as an important prognostic parameter.10 Interestingly, the distribution of different fitness categories at diagnosis was similar to that shown in large B-cell lymphoma10: fit cases represented 50%, unfit cases 36%, and frail cases 14% of our population. Fitness status was reassessed at relapse, showing an increased rate of unfitness (41%) and frailness (33%). In our study, fit patients with late-POD had the best outcome, while in the context of unfit/frail cases, the unfavorable prognosis was primarily influenced by fitness status itself rather than time to POD. Overall, we confirmed, in a real-life setting, the results of previous studies5, 13 about the advantage of using BTKi as a standard second-line treatment. Our median PFS-2 with BTKi was slightly inferior compared to the literature,13 but raised to 25 months considering only fit patients. Median PFS-2 with BTKi was significantly longer compared to CIT, but we did not observe an advantage in terms of OS-2. This finding might be partly explained by the effective use of BTKi as third-line therapy in some patients who relapsed or progressed after CIT. Despite advanced age, frequent cardiac comorbidities, and poor fitness status, treatment with BTKi was feasible and relatively safe, with a low percentage of atrial fibrillation (3%), although we acknowledge that this figure could be under-reported in a retrospective study. Discontinuation rates due to PD, intolerance, or other causes in our cohort were in line with the literature.14-16 Our study had several limitations. First, the lack of central pathological review and biological characterization of MCL patients (e.g., TP53 mutations). Second, the retrospective nature of the analysis may not rule out selection bias in patient enrollment; moreover, differences in outcomes may be due to patients' characteristics that influenced the choice of a specific treatment, and to an improvement in PFS over time compared to the performance of CIT in the early years of our series. Third, survival curves in this real-world experience are probably negatively influenced by less effective salvage therapies compared to the new approved treatment for r/r MCL. The therapeutic algorithm of MCL is rapidly evolving, with BTKi being anticipated to first line, either in addition to the backbone of CIT or possibly as its replacement (as suggested by SHINE, ECHO, and ENRICH trials).17-20 This will probably modify treatment of r/r disease, requiring a repositioning of CAR-T and next-generation biological agents. Patient's fitness category emerges as an essential prognostic parameter in the elderly population, even stronger than time to POD, and deserves further investigation to guide therapeutic decisions both in clinical practice and in future trials. The authors are grateful to the patients who participated in the study, their family and caregivers, and the study staff and health care providers. The authors thank Mattioli Health 1885 srl for the project support and Dr Martina Montagnoli (Verona) for the precious administrative assistance. Francesca Maria Quaglia: Conceptualization; investigation; writing—original draft; methodology; writing—review and editing; formal analysis; project administration; data curation; supervision; visualization. Annalisa Arcari: Conceptualization; investigation; writing—original draft; methodology; writing—review and editing; formal analysis; project administration; data curation; supervision; visualization. Lucia Morello: Investigation; writing—original draft; writing—review and editing; methodology; formal analysis; data curation; visualization. Luigi Marcheselli: Formal analysis; software. Alessandra Tucci: Investigation; writing—review and editing. Chiara Pagani: Investigation; writing—review and editing. Valentina Bozzoli: Investigation; writing—review and editing. Greta Scapinello: Investigation; writing—review and editing. Francesco Piazza: Investigation; writing—review and editing. Vittorio Ruggero Zilioli: Investigation; writing—review and editing. Cristina Muzi: Investigation; writing—review and editing. Benedetta Puccini: Investigation; writing—review and editing. Benedetta Sordi: Investigation; writing—review and editing. Maria Chiara Tisi: Investigation; writing—review and editing. Nicolò Rampi: Investigation; writing—review and editing. Alessia Castellino: Investigation; writing—review and editing. Sara Veronica Usai: Investigation; writing—review and editing. Jari Paternoster: Investigation; writing—review and editing. Emanuele Cencini: Investigation; writing—review and editing. Simone Ferrero: Investigation; writing—review and editing. Chiara Consoli: Investigation; writing—review and editing. Marco Basso: Investigation; writing—review and editing. Elisa Lucchini: Investigation; writing—review and editing. Elsa Pennese: Investigation; writing—review and editing. Maria Elena Nizzoli: Investigation; writing—review and editing. Michele Spina: Investigation; writing—review and editing. Luca Pezzullo: Investigation; writing—review and editing. Maria Stella De Candia: Investigation; writing—review and editing. Rita Tavarozzi: Investigation; writing—review and editing. Mauro Krampera: Investigation; writing—review and editing. Carlo Visco: Conceptualization; writing—review and editing; data curation; supervision; methodology; project administration; investigation. Not applicable. The authors declare no conflicts of interest. We planned, conducted, and reported our research in accordance with the Helsinki Declaration, as revised in 2013 (www.wma.net/policies-post/wma-declaration-of-helsinki-ethical-principles-for-medical-research-involving-human-subjects/). Our research was approved by the Institutional Review Board. Living patients had signed a specific informed consent to the study. For patients lost to follow-up, every reasonable effort was made to trace them, via telephone, email, or by contacting their closest family members. Deceased patients were also included, as excluding them would have altered the study results with major bias. Even for deceased patients, the closest relatives were contacted to inform them of the present study. The authors declare that they have received funding for the publication from the University of Verona, Italy. We do not publish identifying information, including names, initials, or hospital numbers, in any part of our research. We did not use photographs of any part of the body that could identify our patients, so we do not need written consent from the patient (or relatives if deceased). The data are not publicly available due to privacy or ethical restrictions. The data that support the findings of this study are available on request from the corresponding author. 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.