Another disappointment in treating relapsed, refractory high-risk aggressive B-cell lymphomas.

Abstract

Optimising outcomes for patients with relapsed or refractory diffuse large B-cell lymphoma (DLBCL) remains one of the greatest challenges still facing haemato-oncologists today. The SCHOLAR-1 analysis pooled trial and retrospective data (n = 636) and served to outline the bleak outcomes [median overall survival (OS) 6·3 months] for many of our patients who fail to respond to initial treatment. Poor outcomes included patients refractory to front-line potentially curative treatment (median OS 7·1 months) or relapsing within 12 months of autologous stem cell transplantation (autoSCT) (median OS 6·2 months) (Crump et al., 2017). Clear progress has been made over recent years in the field of cellular therapy, and recently for relapsed, refractory DLBCL patients. Chimeric antigen receptor (CAR) T-cell therapy harnesses patients’ own T-cell-mediated immunity to induce a CD19-directed DLBCL tumour assault. High initial efficacy and durable responses in a significant minority of patients have been demonstrated in phase-II single-arm trials of patients with relapsed or refractory aggressive B-cell non-Hodgkin lymphoma (NHL) (predominantly DLBCL) (Neelapu et al., 2017; Schuster et al., 2018). However, as the authors of the accompanying manuscript highlight, CAR-T therapy is associated with significant toxicities (Neelapu et al., 2018), the challenge of equitable, timely administration across populations, and a substantial financial burden to health-care systems (Hernandez et al., 2018). Whilst maximising cure in the front-line setting is clearly still the preferred goal, the quest remains to find highly efficacious, non-toxic, and widely applicable therapies in patients failing front-line anthracycline-based immunochemotherapy who have otherwise failed or are unsuitable for autoSCT. Pixantrone is a novel aza-anthracenedione which is structurally similar to anthracyclines and is licenced in relapsed or refractory DLBCL and National Institute for Health and Care Excellence (NICE)-approved in the UK following the PIX301 trial (Pettengell et al., 2012). PIX301 compared 85 mg/m2 of pixantrone on days 1, 8 and 15 of a 28-day cycle for up to six cycles, with the control arm of ‘best available monotherapy’ for patients with aggressive B-cell NHL (primarily DLBCL). Seventy patients were randomized in each arm, including single-agent vinorelbine (16%), oxaliplatin (46%), ifosfamide (18%), etoposide (13%), mitoxantrone (6%) and gemcitabine (1%) in the non-pixantrone arm. End of treatment complete response (CR) and overall response rates (ORR) were superior in the pixantrone arm [ORR 37·1% (CR 20·0%) vs. ORR 14·3% (CR 5·7%); CR P = 0·021; ORR P = 0·003]. This resulted in a modest progression-free survival (PFS) advantage in favour of pixantrone (median PFS by intention-to-treat 5·3 vs. 2·6 months; hazard ratio (HR) 0·60, P = 0·005) but no OS benefit was seen. Patients were eligible for this trial if they were initially anthracycline-sensitive. Eligible patients had a documented response to prior anthracycline-based therapy of at least 24 weeks, introducing a key potential source of bias within the study design. Subsequently, a UK-wide retrospective non-trial cohort series (n = 92) demonstrated more limited efficacy (ORR 24%, CR 10%) of pixantrone monotherapy in a heavily-pretreated [median prior lines 3 (range: 2–9)], high-risk patient group (Eyre et al., 2016). Eighty-five percent had refractory DLBCL and 72% had an international prognostic index of 3–5. The median PFS in this UK cohort following NICE approval was a disappointing 2·0 months [95% confidence interval (CI) 1·5–2·4] and the median OS was 3·4 months. Unsurprisingly, patients with anthracycline-sensitive disease (relapsed >12 months from front-line treatment) had a significantly improved PFS on multivariable analysis (HR: 0·43; 95% CI 0·22–0·82;P = 0·011). In light of these data, we eagerly awaited the results of the PIX306, published in this edition of the British Journal of Haematology (Pettengell et al., 2019). Pixantrone combined with rituximab (PIX-R) was randomised 1:1 against gemcitabine plus rituximab (GEM-R) as the standard-of-care arm in this phase-III multicentre trial of 312 patients with relapsed aggressive B-cell NHL deemed ineligible for autoSCT. Briefly, there were no significant differences in median PFS [7·3 (95% CI 5·2–8·4) months with PIX-R vs. 6·3 (95% CI 4·4–8·1) months with GEM-R (HR: 0·85; 95% CI 0·64–1·14; P = 0·28)] or indeed median OS [13·3 (95% CI 10·1–19·8) months with PIX-R vs. 19·6 (95% CI 12·4–31·9) months with GEM-R (HR: 1·13; 95% CI 0·83–1·53)] between the treatment arms. No patient subgroups could be identified for whom PIX-R was statistically more beneficial in terms of survival. The PFS was longer in the control arm than the investigators expected; the initial statistical calculations assumed a median PFS of 2·8 months with GEM-R. The improved PFS was likely due to the exclusion of patients with refractory DLBCL from an otherwise ‘trial-fit’ population. Ultimately this is a negative trial and showed that PIX-R is broadly equivalent in terms of efficacy and survival to GEM-R in patients with aggressive B-cell NHL enriched for DLBCL. Despite this, the authors are to be strongly congratulated for conducting this large international collaborative effort; one of the very few randomised controlled trials in relapsed, refractory aggressive B-cell NHL and the only published phase-III trial to date in exclusively autoSCT-ineligible aggressive B-cell NHL patients. Despite the disappointing outcome of the PIX306 trial, hope remains elsewhere for our patients. The antibody–drug conjugate polatuzumab vedotin targeting the CD79b component of the B-cell receptor, in combination with bendamustine and rituximab (BR), has recently received Food and Drug Administration (FDA) approval for the treatment of adult patients with relapsed or refractory DLBCL for whom autoSCT is not suitable. Approval was based on a small randomised phase-II trial where this combination showed an improved ORR (PET CR: 40% vs. 18%; P = 0.026), PFS (HR: 0.34; P < 0.0001) and OS (HR: 0.42; P = 0.0023) compared to BR (Sehn et al., 2018). This combination seems to represent a step forward in this setting, although no randomised phase-III trial is planned. We do, however, eagerly await the results of the randomised phase-III POLARIX trial (NCT03274492) comparing front-line polatuzumab vedotin plus R-CHP versus R-CHOP. Ongoing clinical trials in the field of relapsed, refractory DLBCL continue to investigate targeted agents of promise. CD19 monoclonal antibodies such as the Fc-engineered tafasitamab in combination with lenalidomide are undergoing active investigation (Salles et al., 2019). This combination has shown interesting efficacy in a phase-II trial (L-MIND; NCT02399085) of relapsed (not refractory) DLBCL. The antibody is currently being tested in a randomised phase-III clinical trial in combination with bendamustine (B-MIND; NCT02763319) and the final results of both studies are awaited with interest. Bi-specific T-cell engagers such as mosunetuzumab or CD20-TCB (RG6026) engage T cells by co-targeting CD3 and a B-cell surface marker (typically CD19 or CD20) and are a drug class of great potential, with impressive efficacy in early phase-I–II trials (Budde et al., 2018; Dickinson et al., 2019). Antibody–drug conjugates such as loncastuximab tesirine (Lonca), which comprises a humanized anti-CD19 antibody conjugated to a potent pyrrolobenzodiazepine dimer toxin, have displayed early efficacy signals. Lonca at ≥120 μg/kg has substantial anti-tumour activity in relapsed, refractory DLBCL (ORR 43.3%) displaying equivalent efficacy in refractory patients compared to those with relapsed disease (Radford et al., 2019). Finally, the novel combination of rituximab, lenalidomide and ibrutinib has shown encouraging activity (ORR 65%) in relapsed or refractory non-germinal centre type DLBCL in response-evaluable patients (Goy et al., 2019). Despite the genuine interest in these newer agents with novel mechanisms of action, continuing to demand the testing of agents demonstrating efficacy in early-phase single-arm trials in large, well-designed randomised clinical phase-III trials such as PIX306 must remain the focus of the international treating community. Only then will we truly understand the value of novel agents of promise in relapsed or refractory DLBCL.