Marginal zone lymphoma international prognostic index (mzl‐ipi): a prognostic score for the entire spectrum of marginal zone lymphomas. a fil and spore‐mer study

Background: Marginal zone lymphomas (MZL) include extranodal (ENMZL), nodal (NMZL), and splenic (SMZL) subtypes. While MZL subtypes have largely been studied as separate entities, most clinical trials evaluate MZL as a single entity. In this setting, a prognostic score for all MZL would be very useful. Methods: We analyzed patients that were prospectively enrolled in the NF10 observational study with the aim to define a prognostic score for all MZL. We included adult patients with MZL who started a systemic therapy. For the purposes of this study, patients were classified as SMZL, ENMZL and NMZL according to pathologic diagnosis. Patients without a clear pattern of organ involvement were categorized as disseminated MZL (dissMZL). The primary study endpoint was progression-free survival (PFS) which was calculated from the date of treatment start. For validation, we applied the same inclusion criteria and model used for the NF10 study to an independent cohort of patients from the University of Iowa/Mayo Clinic Lymphoma Specialized Program of Research Excellence (SPORE) Molecular Epidemiology Resource (MER). Results: Starting from the NF10 study, we identified 501 eligible patients: 166 SMZL (33%), 197 ENMZL (39%), 60 NMZL (12%) and 78 dissMZL (16%). At presentation, 40% of the patients were >70 years old, 80% were stage III-IV, 31% had elevated LDH, 41% had Hb <12 g/dl, 20% had lymphopenia (absolute lymphocyte count <1 × 109/L), and 14% had platelets <100 × 109/L. After a median follow-up of 61 months, 5-year(y) PFS was 72% (95% CI 68%–76%). In the final multivariate model, elevated LDH, anemia, lymphopenia, thrombocytopenia, and subtype (NMZL or dissMZL) were independently associated with a worse PFS. A prognostic model was then built with those 5 factors, and patients were classified into low (LRG, 0 factors, 27%), intermediate (IRG, 1–2 factors, 57%) and high (HRG, 3+ factors, 16%) risk groups. 5y PFS was 85% for the LRG, 66% for IRG, and 37% for HRG (Figure), with c-Harrell = 0.64 and robust internal validation and calibration. Compared to the LRG, the IRG (Hazard Ratio [HR] = 2.30, 95% CI 1.39–3.80) and HRG (HR = 5.41, 95% CI 3.12–9.38) had inferior PFS. In the validation using the MER cohort of 192 MZL patients, 5y PFS was 57% (95% CI 51–64). Applying the MZL-IPI to the MER, 41(21%), 113 (59%), and 38 (20%) patients were classified as LRG, IRG, and HRG, respectively. The MZL-IPI was associated with PFS (log-rank test p = 0.043; c-Harrell = 0.60, 95% CI 0.55–0.66); compared to the LRG, the IRG (HR = 1.57, 95% CI 0.97–2.54) and HRG (HR = 2.04, 95% CI 1.15–3.62) had inferior PFS. In both the training and the validation studies, MZL-IPI was associated with the best prediction and discrimination performance, as compared to the IPI, FLIPI and MALT-IPI, and was also prognostic for overall survival. Keywords: indolent non-Hodgkin lymphoma, risk models Conflicts of interests pertinent to the abstract S. Luminari Consultant or advisory role: Roche, Jannsen, Incyte, Gilead, Beigene, BMS, Regeneron S. Ferrero Consultant or advisory role: Janssen, EUSA Pharma, Incyte, Clinigen, Italfarmaco Honoraria: Janssen, EUSA Pharma, Servier, Gentili Research funding: Janssen, Gilead, Morphosys, Beigene I. Del Giudice Consultant or advisory role: AstraZeneca B. K. Link Consultant or advisory role: include Genentech/Roche and MEI Inc D. Mannina Consultant or advisory role: GSK, Kyowa Kirin L. Arcaini Consultant or advisory role: Roche, Janssen-Cilag, Verastem, Incyte, EUSA Pharma, Celgene/Bristol Myers Squibb, Kite/Gilead, ADC Therapeutics Honoraria: EUSA Pharma, Novartis Research funding: Gilead Sciences