Heads or tails? Choosing a salvage therapy for relapsed/refractory Hodgkin lymphoma

Keywords::

• antibody–drug conjugate
• brentuximab vedotin
• novel agents
• refractory Hodgkin lymphoma
• salvage combination chemotherapy

Hodgkin lymphoma (HL) is an uncommon malignancy, accounting for approximately 11% of all malignant lymphomas. It is estimated that in 2012, there will be 9060 new HL diagnoses and 1190 deaths in the USA [1]. With current treatment modalities, sustained remission rates can be achieved in up to 80% of patients [2,3]. Since modern therapy is effective in curing most patients, current studies are focusing on reducing acute and long-term toxicities associated with therapy.

However, despite the success of modern therapy, approximately a third of patients with HL will relapse after first-line therapy, and 15% will fail both first- and second-line therapies [4]. In the relapse setting, choosing the best salvage therapy is a challenge, again, given concerns for long-term toxicities as well as for appropriate efficacy. Salvage combination chemotherapy followed by high-dose chemotherapy with autologous stem cell transplantation (HDT/ASCT) is considered the standard-of-care for relapsed/refractory HL and can induce durable remissions in up to 50% of patients [5]. The importance of chemosensitivity and the type of response before HDT/ASCT in predicting the success of this approach has been underlined by several studies [6–11]. A study by Moskowitz et al. reported significantly better event-free survival, progression-free survival (PFS) and overall survival (OS) after HDT/ASCT for patients responding to second-line therapy (60, 62 and 66%, respectively) when compared with those with poor responses (19, 23 and 17%, respectively; p < 0.001) [12]. Sirohi et al. have reported that patients who were in complete response (CR) or partial response (PR) at the time of HDT/ASCT had better 5-year OS rates (79 and 59%, respectively) when compared with those with disease that was resistant to second-line salvage chemotherapy prior to HDT/ASCT (17%; p < 0.001) [13]. Due to the generally small number of patients and the heterogeneous population enrolled in these studies, analyses comparing different salvage regimens become difficult. To date, there have been no published randomized trials comparing the effectiveness of different conventional chemotherapy salvage regimens. Clinicians have to rely mainly on Phase II data when selecting salvage regimens, and focus is usually given to the potential to induce high response rates (RRs) with low toxicities, providing the ideal scenario to proceed with myeloablative regimens. Traditionally, two primary approaches have been used. Many centers opt for aggressive induction regimens prior to HDT/ASCT, such as carmustine, etoposide, cytarabine and melphalan (mini-BEAM) or dexamethasone, carmustine, etoposide, cytarabine and melphalan (Dexa-BEAM) [8,14], but these regimens have been associated with increased toxicities and a 2–5% risk of treatment-related death. Other centers have used platinum-based regimens, such as ICE (ifosfamide, carboplatin and etoposide) [15], ESHAP (etoposide, solumedrol, cytarabine and cisplatin) [7], DHAP (dexamethasone, cytarabine and cisplatin) [9] or ASHAP (doxorubicin, solumedrol, cytarabine and cisplatin) [16].

Within the past decade, newer salvage combination chemotherapy regimens have been studied for relapsed or refractory HL, and have been shown to induce significant responses, with minimal toxicity. In 2007, Santoro et al. reported on the use of a regimen of ifosfamide, gemcitabine and vinorelbine (IGEV) for relapsed or refractory HL, prior to HDT/ASCT [17]. Ninety one patients were enrolled in this study, with the majority of patients (76.9%) having received only one previous chemotherapy regimen, and all patients having received previous anthracycline-based regimens. Overall RR (ORR) to IGEV was 81.3%, with 49 patients achieving CR (53.8%) and 25 achieving PR (27.5%). Sixty four patients who achieved CR or PR after IGEV proceeded to single (29 patients) or tandem (35 patients) HDT/ASCT, and five patients who responded received non-myeloablative ASCT. The 3-year PFS and OS were 52.98 and 70.03%, respectively. Toxicities from this salvage regimen were mild with a relatively low rate of grade 3 and 4 toxicities and no treatment-related deaths. The most common toxicities were neutropenia and thrombocytopenia. Adequate stem cell collection following induction was observed in more than 80% of patients.

In 2007, Bartlett et al. reported the results of the CALGB 59804 trial [18]. Ninety one patients with relapsed or refractory HL were enrolled in this Phase I/II trial, including patients who received prior HDT/ASCT (40 patients), and received the regimen gemcitabine, vinorelbine and liposomal doxorubicin. ORR for all patients was 70%, with CR in 19% and PR in 51% of patients. For patients who received prior HDT/ASCT, ORR was 75% with 17% CR. For transplant-naive patients, 4-year PFS and OS were 52 and 70%, respectively. Median PFS and OS for the prior transplant group were 8.5 months and 3.5 years, respectively. Incidence of febrile neutropenia with this regimen was 7% and there were no treatment-related deaths in the transplant-naive group.

With all the cited data, clinicians still face a significant challenge in choosing an appropriate second-line regimen for relapsed/refractory HL patients. The lack of comparative prospective trials between regimens force clinicians to decide based on the data of toxicity/response balance and personal experience. Cross-comparison between trials, although not adequate, has been used to guide these decisions also. In our opinion, the use of aggressive second-line induction regimens prior to HDT/ASCT, such as mini-BEAM or Dexa-BEAM, should be reserved for young and fit patients, with significant poor prognostic features (i.e., primary refractory disease, extent of disease and poor sensitivity to chemotherapy), given the higher rates of toxicities arising from these regimens, and even in these settings the risk/benefit ratio should be carefully considered. Although platinum-based regimens have been the traditional ‘go-to’ regimen of choice for clinicians in the USA, current evidence suggests that newer regimens such as IGEV can induce higher CR rates, while minimizing toxicities (in cross-trial comparisons, CR rates for IGEV, ASHAP, DHAP and ICE were 54, 34, 21 and 26%, respectively) [9,12,15–17].

Even with advances in improving CR rates prior to HDT/ASCT and safety standards that reduce the treatment-related mortality of transplant, there is a significant patient population who do not achieve sustained benefits from this treatment modality. Particularly, patients whose disease relapses within 1 year of HDT/ASCT carry a poor prognosis, with a median OS of 1.2 years [5]. Until recently, there was no standard-of-care for salvage treatment in this population. This paradigm has changed, with the antibody–drug conjugate brentuximab vedotin (BV) being approved for treatment of patients with HL whose disease relapsed after HDT/ASCT (or after two multiple-drug chemotherapy regimens in patients who are not eligible for transplant). This antibody–drug conjugate combines a monoclonal antibody targeting CD30 with the antimicrotubulin agent monomethyl-auristatin E, and Phase II data show impressive responses, with ORR at 75% with 34% of patients achieving CR [19]. It is important to underline that this patient population had a very poor prognosis, with disease generally refractory to front-line therapy and a median time-to-relapse after HDT/ASCT of only 6.7 months. These results are encouraging, and current efforts are underway to incorporate BV into earlier lines of treatment, including combining BV with standard combination regimens for newly diagnosed HL patients, and two studies testing BV as salvage therapy prior to HDT/ASCT [101,102].

Future directions for salvage therapy in HL will probably incorporate novel agents into traditional multiagent chemotherapy regimens. Phase I data of bortezomib in combination with ICE have yielded promising remission rates and PFS, and was well tolerated [20]. Recently, Younes et al. reported on the single-agent activity of the histone deacetylase inhibitor panobinostat in patients with relapsed HL after HDT/ASCT [21]. In this trial, reductions in tumor size were seen in 74% of patients, with 27% achieving CR or PR. Median duration of response was 6.9 months and 1-year OS was 78%. A current Phase I trial [103] is being conducted, combining ICE with panobinostat, based on the in vitro evidence of synergy between histone deacetylase inhibitors and chemotherapy. Other novel drugs that were recently shown to be active in HL include the mTOR inhibitor everolimus and the immunomodulatory agent lenalidomide [22,23]. However, given the modest single-agent activity, it is unlikely that these agents will have a significant role in the treatment of HL, unless rational combinations with other novel agents or conventional chemotherapy drugs prove to be more effective.

Financial & competing interests disclosure

B Pro has received honoraria and a research grant from Seattle Genetics. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.