The treatment of relapsed/refractory anaplastic large cell lymphoma expressing the anaplastic lymphoma kinase: a single-center experience

Abstract

Anaplastic large cell lymphoma (ALCL) expressing the anaplastic lymphoma kinase (ALK) (ALK+ ALCL) is a rare type of lymphoma which comprises 10-15% of all non-Hodgkin lymphomas in children and 2–3% in young adults. Relapsed/refractory disease occurs even more rarely (25–40% of all cases). There is as yet no standard treatment for relapsed/refractory ALK+ ALCL. Patients with ALK+ ALCL usually present at advanced stages of the disease with extranodal involvement (skin, soft tissues, bones, lungs, liver, spleen and bone marrow) and B symptoms. ALK-positive ALCL affects males more often than females. There are two morphological variants: the common type (65% of cases) and the non-common type which is associated with a poorer prognosis. ALK+ ALCLs are often associated with t(2;5) and t(1;2), resulting in the formation of the NPM-ALK and the TPM3-ALK fusion proteins, respectively. Data about the treatment of relapsed/refractory ALK+ ALCL are limited. Earlier, targeted therapies (brentuximab vedotin (BV), ALK inhibitors) and risk-adapted chemotherapy followed by hematopoietic stem cell transplantation (HSCT) for remission consolidation were shown to be highly effective. A total of 15 patients with relapsed/refractory ALK+ ALCL were treated at the R.M. Gorbacheva Research Institute starting from 2002. Fourteen (93%) patients had ALK-positive ALCL of common morphology and one (7%) patient had the non-common variant (histiocytic). The study was approved by the Independent Ethics Committee and the Scientific Council of the Pavlov University. The expression of CD3 on tumor cells was assessed (CD3 positive: n = 4 (27%), CD3 negative: n = 8 (53%), no data: n = 3 (20%). The median age at the diagnosis was 26 years (11 months– 37 years). The median follow-up from the diagnosis was 9 years (1–19 years). Nine (60%) patients were aged > 18 years and six (40%) patients were aged < 18 years. There were 10 (67%) males and 5 (33%) females. At onset, 2 (13%) patients were diagnosed with early-stage disease (stage II), while the others were diagnosed with advanced-stage disease: 2 (13%) patients had stage III disease and 11 (74%) had stage IV disease. Staging was performed according to the St. Jude staging system (in children) and the Ann Arbour staging classification (in adults). Thirteen (86%) patients had extranodal involvement. Four (27%) patients had refractory disease (progression within the first three months or the absence of complete remission after the first-line treatment) and the rest 11 (73%) patients had recurrent ALK-positive ALCL. Six patients developed early relapse (< 12 months after remission was achieved); 5 patients had late relapse (after > 12 months of remission); local (1 site) and systemic relapses were diagnosed in 7 and 4 patients, respectively. Our patients received from 2 to 7 lines of treatment (the median is 4). In the first line of therapy, the patients were treated according to NHL-BFM based regimens (n = 9; 60%), the CHOP (n = 5; 33%), and the HyperCVAD (n = 1; 7%) protocols. In the second line of therapy, 8 (53%) patients were treated according to NHL-BFM based regimens; 2 (13%) patients were treated with GDP; 1 (7%) patient received DHAP chemotherapy; 1 (7%) patient received a combination of methotrexate and vinblastine (MTX + V); 1 (7%) patient received bendamustine as a single agent. Two (13%) patients were treated with chemotherapy in combination with targeted drugs (GDP + BV, n = 1; NHL-BFM + crizotinib, n = 1). As a third or subsequent line of treatment, the patients received a variety of chemotherapy regimens (n = 5; 33%) and chemotherapy in combination with targeted drugs (n = 10; 67%). Five (33%) patients underwent ALK-inhibitor therapy (crizotinib (n = 4) and ceritinib (n = 1)). Seven (46%) patients were treated with BV (BV as a single agent (n = 4) and BV + chemotherapy (n = 3)). The median number of treatment lines before autologous HSCT (auto-HSCT) and allogeneic HSCT (allo-HSCT) was 2 (2–3) and 3 (3–4), respectively. Auto-HSCT was carried out in 11 (73%) cases. Nine (60%) patients underwent allo-HSCT (from a matched unrelated donor (n = 6), from a matched related donor (n = 2), and from a haploidentical donor (n = 1)). One patient received NK cells from a haploidentical donor as maintenance. In 5 (33%) cases, alloHSCT was carried out following auto-HSCT. The conditioning regimens (CR) used for auto-HSCT included BEAM (carmustine – 300 mg/m2, etoposide – 800 mg/m2, Cytosar – 1600 mg/m2, melphalan – 140 mg/m2) – in 5 (45%) patients; BeEAM (bendamustine – 320 mg/m2, etoposide – 800 mg/m2, Cytosar – 1600 mg/m2, melphalan – 140 mg/m2) – in 5 (45%) patients; and BuCy (Cyclophosphan – 100 mg/kg, busulfan – 14 mg/kg) – in 1 (10%) case. Seven (78%) patients undergoing allo-HSCT received the FluBenda conditioning regimen (fludarabine – 90–150 mg/m2, bendamustine – 390 mg/m2) and post-transplant Cyclophosphan and calcineurin inhibitors for the prevention of graft-versus-host disease (GVHD) (n = 7; 78%); one (11%) patient received the FluMel regimen (fludarabine – 150 mg/m2, melphalan – 140 mg/m2) and CsA/MTX (Cyclosporin А, methotrexate) for GVHD prevention; and in 1 case (11%) data on the RC and GVHD prophylaxis were missing. Overall response to the second line of treatment was achieved in 10 (67%) patients, with complete response observed in 7 (47%) cases, and partial response –in 3 (20%) cases. Five out of the 7 patients treated with BV during different lines of therapy managed to achieve complete response. Four out of the 5 patients who had undergone treatment with ALK inhibitors, demonstrated complete response. The 10-year overall survival (OS) rate of the study patients reached 90% (95% confidence interval (CI) 47–99). The 10-year progression-free survival (PFS) rate after the second line of treatment was 39% (95% CI 13–64). The 10-year OS and PFS rates after auto-HSCT were 100% and 35% (95% CI 8–64) respectively. The 5-year OS following allo-HSCT was 85% (95% CI 33–98), while PFS was 60% (95% CI 19–85). Four out of the 11 patients who had undergone auto-HSCT relapsed, and 2 patients progressed. Median time to relapse/progression was 8 (6–27) months. Three out of the 9 patients who had been treated with allo-HSCT ended up relapsing (median time: 8 (6–17) months). Two patients achieved repeated remission (in one case, it was the result of treatment with ceritinib, while in the other case it became possible after the resection of the lesion, radiotherapy and prescription of crizotinib), and 1 study patient died as a result of disease progression 17 months after allo-HSCT. The 6 patients who had achieved complete remission before undergoing allo-HSCT, are still in CR. Five out of the 9 patients developed grade I–II acute GVHD with skin involvement but did not show any signs of chronic GVHD. The observed complications of chemotherapy and auto-HSCT were standard and manageable and were not the focus of attention in this study. Taking into account the high probability of developing resistance to ALK inhibitors and the high risk of relapse after treatment with BV, targeted therapy should be used to prepare patients for HSCT. The use of ALK inhibitors and BV in our study led to repeated remission in 80% and 71% patients respectively. It was demonstrated that in the majority of cases even heavily pretreated patients with ALK+ ALCL can be cured (which is not the case with other non-Hodgkin lymphomas), especially if allogenic HSCT (allo-HSCT) is carried out. Still, we think that auto-HSCT can also be considered for remission consolidation since OS rates following auto-HSCT and allo-HSCT are comparable (100% and 85%). Moreover, auto-HSCT can also be a valid option in the absence of a fully matched donor, i.e. when only an alternative (haploidentical or partially matched) donor is available, since the use of haploidentical HSCT in ALCL patients has not been studied well enough yet. Further research on the matter is warranted. It was demonstrated that a non-myeloablative conditioning regimen before allo-HSCT (FluBenda) could be opted for in patients with relapsed/refractory ALK+ ALCL and that it was similarly effective as myeloablative RC when compared with a historical control group. Disease status before HSCT was proven to have a statistically significant influence on PFS (the best prognosis was associated with complete remission). At the same time, other factors did not impact the prognosis. This may be explained by the relatively small number of patients included in the study. Relapsed/refractory ALK+ ALCL is a disease with a relatively good prognosis even in heavily pretreated patients. Targeted therapy is a very important and effective step in preparation for HSCT. Allo-HSCT is more effective than auto-HSCT but the latter can also be considered a valid therapeutic option.