Innovative therapies based on molecular orientation in patients with relapse and refractory diffuse large B ‐cell lymphoma: Results of LNH‐EP1 study
Year: 2021
Type: article
Abstract: With the standard of care (SOC), namely R-CHOP, 30%–35%1 of diffuse large B-cell lymphoma (DLBCL) patients will present a refractory or relapsing (RR) disease and only 15% of them will experience a prolonged response after intensive salvage.2 Management of patients with RR DLBCL continues to represent a major therapeutic challenge and clinical trials evaluating innovative drugs could give access to new options of treatment for these patients. Some studies have demonstrated the clinical benefit of molecular orientation to guide anticancer therapies in the setting of solid tumors.3, 4 In the context of lymphoma, targeting specific molecular aberrations might improve outcomes of patients. For example, targeting EZH2 mutations with EZH2 inhibitors, or CD79B mutations with BTK inhibitors. However, a personalized molecular therapeutic approach in the treatment of relapsed or refractory (RR) lymphoma has not been yet evaluated in clinical studies, as the majority of the early phase trials evaluating targeted agents did not implement molecular enrichment for patient's selection. In the present single-center study, we evaluated the clinical benefit of targeted NGS in patients with RR DLBCL to guide innovative treatment decision. Study design and patient's selection (see Supplementary section for complete methods): The LNH-EP1 study was conducted in two consecutive steps. The step one consisted in the tumor molecular characterization (panel of 44 genes, Table S1) of patients ≥18 years with RR DLBCL after at least one line of systemic treatment (2015–2019) (study flowchart, Figure S1). Of the 103 patients included, after tumor analysis and DNA extraction, a molecular portrait was successfully obtained in 88 (85%). The LNH-EP1 step two study consisted in the analysis, for each patient's, of molecular alterations by an institutional molecular tumor board, to assess the possibility of molecular orientation to guide innovative therapy (based on the ESCAT classification5). Eligibility criteria for LNH-EP1 step two were: molecular analysis from the LNH-EP1 step 1, age 18+ years, histological confirmation of RR DLBCL, PS score 0–2, 2+ lines of systemic prior treatment, a measurable lymphoma disease according to Cheson 2016 criteria,6 patient candidate for innovative therapy, and no other standard treatment available. Within the 88 patients included in step one, 30 were not eligible for step two and received treatment according to the best standards of care (supplementary methods). Fifty-eight eligible patients were included and were further classified in two groups: the first one named “molecularly oriented” (MO, N = 15), in which patients received a treatment based on the presence of a given molecular abnormalities that was considered targetable; and the second one, “non-MO group” (N = 43), in which patients were treated with an innovative therapy or enrolled in a clinical trial but not based on a precise molecular abnormality. Outcome data (response rate, PFS and OS) of MO and non-MO group were then compared to analyze the clinical impact of molecular orientation decision. Note, DNA extraction, panel gene sequencing, mutation calling pipeline and analysis as well as statistical methods are presented in supplementary data. Clinical characteristics of the 88 patients included in step one with successfully obtained tumor molecular portrait at the time of molecular characterization are described in Table S2. Patients had a mean age of 65 years (27–87 years), and received a median of two lines of previous systemic therapies (range 1–9). Sixty-three patients (63/88, 71%) had a history of primary refractory disease (RR within a year from first line). A total of 304 mutations were identified (median/patient of 3, range 0–11, three patients presenting no alteration within the 44 genes panel). So, TP53, CREBBP, KMT2D, PIM1 and BCL2 were the most recurrently mutated genes with an incidence of 43%, 33%, 28%, 25% and 23% respectively (Tables S3 and S4). Within these 88 patients, 58 were included in the LNH-EP1 step two study and 30 did not meet the step two eligibility criteria, due to first relapse (N = 18, and candidate to SOC second line therapy), second relapse, and eligibility for standard high dose therapy (N = 7). Or, not candidates for innovative therapy [N = 5: CNS involvement (N = 1), no slot available in clinical trial and screening failure (N = 4)]. These 30 patients were treated with the best standard of care (Figure S1). The majority of these 30 patients were in first (18/30, 60%) or second relapse (7/30, 23%) at time of molecular analysis and therefore eligible for intensive salvage followed by ASCT (median number of lines of previous therapy one, range 1–5). Fourteen (14/30, 47%) reached a response to salvage (CR or PR). At time of analysis, after a median follow-up of 35 months (1–56 months) since salvage initiation, 10/30 patients were alive in CR with a 2 years OS estimated at 29% (95 CI 0.16–0.52). Fifty-eight patients were eligible and included in the LNH-EP1 step two, 15 treated within MO therapy arm and 43 within non-MO arm. In these 58 patients, the median lines of previous therapy was three (range 2–9). For the 15 MO patients, molecular targets included: MYD88 mutations (N = 6/15, 40%), EZH2 mutations (n = 3/15, 20%), CD79B mutations (N = 3/15, 20%), ARID1A mutations (N = 1/15, 7%), BCL7A mutations (N = 1/15, 7%), BRAF mutations (N = 1/15, 7%) (Table S5). The remaining 43 patients in the non-MO group received the best innovative therapy based on the physician decision: epigenetic modifiers (EZH2, PRMT5 or EED inhibitors, N = 18), kinase inhibitors (SYK, PI3K, BTK, N = 4), BCL2 inhibitors (N = 2), immune-oncology agents (such as anti PD1 and/or anti-CTLA4, N = 9), anti CD19-CD3 bi-specific (N = 1), or cereblon modulators (N = 9). Patients' characteristics in the MO group and no-MO groups are presented in Table S2. At time of molecular characterization, patients in the MO group had a mean age of 59 years versus 68 years in the no-MO group (p = 0.08). The MO patients had received the same number of lines of previous therapies compared to non-MO group [median 4 (2–6) vs. 3 (2–9), p = 0.67], including 27% versus 12% of prior ASCT (p = 0.17). Furthermore, 67% of the MO patients were primary refractory versus 72% in the no-MO group (p = 0.69). Treatment Response based on molecular orientation strategy: Responses in the 15 MO patients are presented in Table S6: the ORR was 60% (9/15) and CR rate 25%. The ORR and CR were significantly higher in the MO group versus non-MO group (60% ORR and 25% CR in the MO group as compared to 21%, with 7% CR in the non-MO group, p = 0.02, Table S7). A greater number of patients experienced a progressive disease in the non-MO group than in the MO group (26% vs. 56%, p = 0.03). Outcome: After a median follow up 38 months (3–72 months, since innovative treatment initiation) for the living patients, 2-year OS estimate was 53% (95% CI: 0.32–0.85) in the MO group versus 11% in the non-MO group (95% CI: 0.05–0.28); 2-year PFS 20% (95 CI: 0.07–0.55) in the MO group versus 2% in the non-MO group (95% CI: 0.00–0.16). In univariate prognostic analysis (Table S8), none of the genes selected for treatment decision were associated with a prolonged PFS or OS. These data suggest that the bias related to molecular selection, if existing, are not in favor of the MO group. Furthermore, within the clinical factors only decreased lymphocyte count (p = 0.007, HR 9), history of indolent lymphoma (p = 0.048, HR = 0.56) and non-MO treatment strategy (p = 0.013, HR = 2) were significantly associated with a shorter PFS; older age (p = 0.001, HR = 2.7), decreased lymphocyte count (p = 0.015, HR = 2.1), history of indolent lymphoma (p = 0.03, HR = 0.49) and non-MO treatment strategy (p = 0.022, HR = 2.5) were significantly associated with shorter OS (Table S8). In a multivariate model including these four variables, non-MO treatment and lymphocyte count decrease remained independently associated with a shorter OS (Figure 1 and Table S9), and lymphocyte count with shorter PFS. This study investigated the potential outcome benefit of precision-medicine approach in RR DLBCL patient's management. This field has been studied in patients with “hard to treat” solid cancers.4 In our study, 15/58 patients (26%) with a molecular portrait performed before targeted treatment initiation had an innovative therapy based on molecular orientation. This corresponded to the incidence in other solid oncology studies (199/843, 23% in MOSCATO trial4), indicating that in the current state of knowledge, a limited number of patients could benefit from molecularly oriented therapy. In our study, patients treated with molecular orientation approach tend to achieve a better outcome, which might be prone to bias in this real-life setting. Indeed, we acknowledge some limitations related to the number of included patients, and bias related to selection of the patients with specific molecular alterations and treatment, available at the time of analysis (either innovative in the setting of clinical trials, but also the absence of CAR-T cells therapy routinely available). However, none of the genes selected for treatment decision were associated with a prolonged PFS or OS, suggesting that potential bias related to molecular selection are not in favor of the MO group. In conclusion, we provide evidence for the feasibility of performing molecular characterization of tumors in daily clinical practice in RR DLBCL to guide early phase clinical trial enrollment. The authors thank the CALYM institute for financial support. JC.S: share holder: AstraZeneca, Relay Therapeutics, Gritstone Oncology; Board of directors: Hoopika Pharmaceuticals; was a full-time employee of AstraZeneca between Sept 2017 and December 2019. A.V: is a full-time employee of AstraZeneca. C.S, J.M.M, C.B, P.R.M, A.G, A.H, R.B are Principal/sub-Investigator of Clinical Trials for Abbvie, Adaptimmune, Aduro Biotech, Agios Pharmaceuticals, Amgen, Argen-X Bvba, Arno Therapeutics, Astex Pharmaceuticals, Astra Zeneca Ab, Aveo, Basilea Pharmaceutica International Ltd, Bayer Healthcare Ag, Bbb Technologies Bv, Beigene, Blueprint Medicines, Boehringer Ingelheim, Boston Pharmaceuticals, Bristol Myers Squibb, Celgene Corporation, Chugai Pharmaceutical Co, Clovis Oncology, Cullinan-Apollo, Daiichi Sankyo, Debiopharm, Eisai, Eisai Limited, Eli Lilly, Exelixis, Faron Pharmaceuticals Ltd, Forma Tharapeutics, Gamamabs, Genentech, Glaxosmithkline, H3 Biomedicine, Hoffmann La Roche Ag, Imcheck Therapeutics, Innate Pharma, Institut De Recherche Pierre Fabre, Iris Servier, Janssen Cilag, Janssen Research Foundation, Kura Oncology, Kyowa Kirin Pharm. Dev, Lilly France, Loxo Oncology, Lytix Biopharma As, Medimmune, Menarini Ricerche, Merck Sharp & Dohme Chibret, Merrimack Pharmaceuticals, Merus, Millennium Pharmaceuticals, Molecular Partners Ag, Nanobiotix, Nektar Therapeutics, Novartis Pharma, Octimet Oncology Nv, Oncoethix, Oncopeptides, Orion Pharma, Ose Pharma, Pfizer, Pharma Mar, Pierre Fabre, Medicament, Roche, Sanofi Aventis, Seattle Genetics, Sotio A.S, Syros Pharmaceuticals, Taiho Pharma, Tesaro, Xencor. C.S: consulting for Incyte. J.M.M: personal fees (lectures): Roche, AstraZeneca; NON-FINANCIAL SUPPORT (Drugs, equipment supplied by the entity, travel paid by the entity, writing assistance, administrative support, etc.): AstraZeneca, Roche, Novartis, Gilead, Celgene, Bristol-Myers Squibb, GSK. C.B.: grant and consulting from Bristol-Myers Squibb. Appendix S1. Supporting Information. Table S3 Incidence of mutations within the 88 patients Table S4 Listing of mutations (excel file) 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.
Source: American Journal of Hematology
DOI: 10.1002/ajh.26280
Authors: Clémentine Sarkozy, Jean‐Marie Michot, Cyril Quivoron, Valérie Camara‐Clayette, Alina Danu, Julien Lazarovici, Yolla El‐Dakdouki, Khalil Saleh, David Ghez, Julien Rossignol, Capucine Baldini, Julia Arfi‐Rouche, Patricia Romano‐Martin, Lambros Tselikas, Andrea Varga, Anas Gazzah, Antoine Hollebecque, Ludovic Lacroix, Rastislav Bahleda, Hélène Lecourt, Veronique Vergé, Wasilla Rahali, Sophie Cotteret, Jean‐Charles Soria, Peggy Dartigues, Christophe Massard, Vincent Ribrag
Institutions: Inserm, Institut Gustave Roussy, Fondation Gustave Roussy
Language: English
Subfield: Pathology and Forensic Medicine
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W3173840596