Improving outcomes after autologous transplantation in relapsed/refractory Hodgkin lymphoma: a European expert perspective

The most widely investigated strategy to improve post-ASCT outcomes is the optimization of pre-ASCT salvage chemotherapy. A variety of regimens have been studied, including platinum-based dexamethasone, cytarabine, and cisplatin/doxorubicin, methylprednisolone, high-dose cytarabine, and cisplatin/etoposide, methylprednisolone, high-dose cytarabine, and cisplatin (DHAP/ASHAP/ESHAP [20‐24]), ifosfamide-based (MINE/ICE/IVE/IVOx [25‐29]), and gemcitabine-based combinations (GVD/IGEV/GDP/GemOx/BeGEV) [30‐34]. Most of these studies evaluated the efficacy of salvage chemotherapy in terms of objective responses and the impact on OS and post-transplant PFS; reports of the impact on stem cell mobilization, stem cell quality, and stem cell transplantation (SCT) rates are inconsistent (Table 1). Objective response rates (ORRs) ranged from 61% with GVD (in ASCT-naïve patients) to 88/89% with ICE and DHAP [20, 26, 27], and complete response (CR) rate was as low as 17% with GDP and as high as 73% with BeGEV [32, 34]. PFS was 78% with ESHAP in patients who achieved CR, but 16% in those who achieved partial remission (PR) [24]; ICE treatment resulted in a PFS of 70% [35], but only 53 and 62% with IGEV and BeGEV, respectively [31, 34]. EFS was 36% with ASHAP and 68% with ICE [22, 26]. OS with ESHAP was estimated at 35% at 3 years in one study and 73% at 5 years in another study [23, 24]. However, the absence of prospective randomized studies comparing the regimens makes it impossible to reach conclusions regarding the superiority of particular combinations. This is reflected in the current European Society for Medical Oncology (ESMO) and National Comprehensive Cancer Network (NCCN) treatment guidelines, which do not recommend any specific salvage therapy regimen for patients with R/R HL [10, 11, 36]. Instead, appropriate therapy is generally selected based on patient-related factors, familiarity of the treatment center with particular regimens, treatment as in-patients versus out-patients, and the toxicity profile of each salvage regimen – patients with concomitant coronary, pulmonary, or renal diseases may require different salvage management.

Patients who achieve a response, particularly a CR, with conventional salvage chemotherapy prior to ASCT, are likely to have an improved clinical outcome compared with patients who have partially or totally chemo-resistant disease at relapse. Moskowitz et al. reported an analysis of long-term outcomes in 75 patients demonstrating significant improvements (p < 0.001) in EFS (60% vs 19%), PFS (62% vs 23%), and OS (66% vs 17%) in patients who responded to standard-dose second-line therapy after relapse, compared with those who had a poor response, respectively [37]. Similarly, in a prospective analysis of 195 patients treated over a 20-year period, patients who had achieved a CR with conventional salvage chemotherapy and who were still in CR at the time of ASCT had a 5-year OS of 79%, which dropped to 59% for those in PR, and 17% for those with resistant disease (p < 0.0001). Corresponding 5-year PFS rates were 69% versus 44% versus 14% (p < 0.0001) [38]. More recently, in a phase II study of 97 patients, those who proceeded to SCT with a positron-emission tomography (PET)-negative status achieved an EFS of > 80%, compared with 29% in PET-positive patients [35]. Collectively, these studies demonstrate that the aim of any modern salvage therapy is to produce a deep remission and PET-negative status prior to undergoing HDC-ASCT. However, ASCT should not be solely withheld due to lack of PET-negativity [39].

The GHSG have investigated the concept of escalated pre-ASCT chemotherapy, delivered as sequential-HDC, compared with standard HDC, in an effort to improve treatment outcomes for patients with R/R HL receiving HDC + ASCT [40]. In the GHSG study of standard versus intensified BEAM-ASCT following DHAP in patients with relapsed HL (n = 241), there were no significant differences in FFTF (p = 0.56) or OS (0.82) between the study arms. Toxicity was considerably higher in the intensified arm, with increased rates of grade 3/4 adverse events (AEs), although this did not translate into increased mortality.

Patients who achieve a CR following induction chemotherapy are highly likely to respond to post-relapse interventions [31]. Previous studies in patients with HL who have experienced multiple relapses have demonstrated that bendamustine monotherapy has promising activity, with patients who were ineligible for ASCT, or for whom ASCT had failed, achieving CR rates of 25–35% [41‐43]. In an open-label phase II study, ORR was 83% (CR = 73%) following 4 cycles of BeGEV as induction therapy before ASCT in 43/49 patients. Two-year PFS and OS rates in the overall patient population were 62.2 and 77.6%, respectively, and 80.8 and 89.3% among patients who underwent ASCT [34].

Brentuximab vedotin is currently approved as a monotherapy for adult patients with R/R CD30-positive cHL following ASCT or following ≥ 2 prior therapies when ASCT or multi-agent chemotherapy is not a treatment option [44]. The pivotal phase II SG035–0003 study of brentuximab vedotin (1.8 mg/kg every 3 weeks [Q3W]) after failed ASCT in 102 patients with R/R HL, reported median OS of 22.4 months and median PFS of 5.6 months at the primary analysis [45]. With prolonged follow-up (median 35.1 months) estimated 5-year OS and PFS rates were 41 and 22%, respectively [46]. Corresponding OS and PFS rates amongst the 33% of patients who achieved a CR were 64 and 52%, and the median response duration was not reached. The prospective phase IV C25007 study evaluated brentuximab vedotin (1.8 mg/kg Q3W) in 60 R/R HL patients who were unsuitable for ASCT or multi-agent chemotherapy [47]. The ORR was 50% (CR = 12%), 47% of patients were bridged to ASCT, and the estimated 12-month OS was 86%, thus enabling patients with high-risk disease to receive ASCT, even if they had a suboptimal response to frontline treatment or chemotherapy/radiotherapy-based salvage.

Lately, brentuximab vedotin has been assessed alone and in combination with conventional chemotherapy-based regimens for salvage therapy prior to ASCT in several studies (Table 2). A phase II study assessed 4 cycles of standard-dose, single-agent brentuximab vedotin (1.8 mg/kg Q3W) as salvage treatment following induction therapy with doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) and/or bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone (BEACOPP) [52]. In 37 patients, ORR was 69% and CR rate was 35%; 89% of patients were able to proceed to ASCT, either with or without additional chemotherapy.

Encouraging results with PET-adapted salvage therapy have been reported by Moskowitz et al. [16] in a phase II open-label study to assess the efficacy of 2 cycles of brentuximab vedotin-based salvage (1.2 mg/kg on days 1, 8, and 15) in patients with R/R HL who had failed one previous doxorubicin-containing regimen. Following the first 2 cycles of brentuximab vedotin, 27% of patients were PET-negative and proceeded directly to ASCT and 69% initiated augmented ifosfamide, carboplatin, and etoposide (aICE) [16]. Seventy-six percent of patients achieved PET-negativity prior to HDC/ASCT, thus maximizing the potential for improved post-ASCT outcomes. Patients treated with 3 cycles of brentuximab vedotin also achieved similar rates of PET-negativity (30%) and proceeded directly to ASCT, whilst the remainder received either aICE or ICE [39]. Prior to transplant, 80% of patients had achieved PET-negativity with either brentuximab vedotin alone or combined with aICE/ICE, and with 2 years’ follow-up EFS was 82%.

Brentuximab vedotin combined with standard ICE has also shown promise in phase I/II clinical studies [50, 51]. In the NCT02227199 study, 20/23 evaluable patients previously treated with ABVD achieved a PET CR per investigator review following brentuximab vedotin + ICE; 70% were in PET CR per independent central review [50]. Final data from the Lymphoma Academic Research Organisation phase I/II study showed that 69% (n = 27/39) of patients who received brentuximab vedotin (1.8 mg/kg) plus ICE achieved a complete metabolic response (CMR) and 26% achieved a partial metabolic response. Twenty patients in CMR went on to receive ASCT [51]. Neither study reported any unexpected toxicities.

Two brentuximab vedotin + platinum-based chemotherapy combinations appear particularly promising: brentuximab vedotin + DHAP (B + DHAP), and brentuximab vedotin + ESHAP (BrESHAP). In the phase II BRaVE study, the high metabolic CR rate of 79% achieved with B + DHAP suggests that this approach is worth investigating further [48]. Moreover, the Grupo Español de Linfomas y Trasplantes de Médula Ósea (GELTAMO) study further demonstrated the potential for brentuximab vedotin combinations with chemotherapy in the pre-ASCT setting. After approximately 27 months of follow-up, 74% of patients were FFTF, PFS was 71%, and OS was 91% [49].

The combination of brentuximab vedotin (1.8 mg/kg Q3W) and bendamustine (90 mg/m² days 1–2 Q3W) was highly active as first salvage therapy for patients with R/R HL in a phase I/II study involving 55 patients [53]. ORR was 93% and CR was 74% after a median of 2 cycles of therapy.

Interim results of another phase I/II study have also demonstrated the utility of 4 cycles of a brentuximab vedotin (1.8 mg/kg Q3W) combined with nivolumab (3.0 mg/kg Q3W) combination as initial salvage therapy for patients with R/R HL [54]. ORR in all treated patients (n = 61) was 82%, and CR was 61%.

In summary, single-agent brentuximab vedotin as a salvage therapy shows activity that is not that different to that seen with classical chemotherapy regimens, albeit with a lower systemic toxicity, although these have not been compared in a prospective, randomized phase III study. Clinical studies of brentuximab vedotin combined with chemotherapy build on the demonstrated efficacy of brentuximab vedotin monotherapy, with high CR rates observed with combination therapy, and with no new safety signals. Again, these observations need to be confirmed in prospective, randomized phase III studies, such as the BRESELIBET study (NCT04378647) that was opened to enrolment in June 2020. Despite the low patient numbers in these studies, brentuximab vedotin plus chemotherapy regimens are frequently used as salvage therapy in the clinic.

Consolidation therapy is administered to patients who may have undetectable residual lymphoma and those who may already be cured. Therefore, the optimal consolidation therapy should have a favorable/low toxicity profile that does not impair hematopoietic and immunological recovery post-ASCT, with minimal impact on non-lymphoid/hematopoietic tissues. As biologic treatment options, such as brentuximab vedotin, are less toxic than traditional cytotoxic chemotherapy, they may be an ideal treatment option in this setting.

The utility of brentuximab vedotin as early consolidation therapy after ASCT in patients with high-risk HL was explored in the phase III AETHERA study (see Table 3) [65]. Patients enrolled in AETHERA (n = 329) had one of the following pre-ASCT risk factors for relapse: primary refractory HL (i.e., CR not achieved with frontline chemotherapy), relapsed HL with a duration of initial remission < 12 months, or extranodal involvement at the start of pre-ASCT HDC. After a median of 30 months’ follow-up, PFS per independent review was significantly longer with brentuximab vedotin compared with placebo (median PFS of 42.9 vs 24.1 months [HR 0.57; 95% CI 0.40–0.81]; p = 0.0013) [65]. PFS benefit was consistent across pre-specified subgroups, including patients with primary refractory HL and those who relapsed < 12 months after frontline therapy. After 5 years of follow-up, brentuximab vedotin continued to demonstrate sustained PFS benefit (HR 0.52; 95% CI 0.38–0.72), with 5-year per investigator PFS rates of 59% versus 41% with placebo, implying that a considerable proportion of these patients might have been cured [17]. This long-term PFS benefit with brentuximab vedotin was more pronounced in patients with higher numbers of pre-ASCT risk factors, which included relapse within < 12 months or refractoriness to frontline therapy, best response of PR or stable disease (SD) to most recent salvage therapy, extranodal disease at pre-ASCT relapse, B-symptoms at pre-transplant relapse, or ≥ 2 prior salvage therapies. An OS analysis has not yet been performed, as the planned number of events has not been reached. Brentuximab vedotin consolidation was well tolerated in the AETHERA study, with 93 of 167 patients (56%) experiencing grade 3/4 AEs [65]. Sixty-seven percent of brentuximab vedotin-treated patients experienced some level of peripheral neuropathy, 90% of which had either completely resolved or decreased in severity after 5 years’ follow-up [17].

The results of the AETHERA study are reflected in the HL treatment guidelines developed by ESMO and NCCN, and the lymphoma post-ASCT maintenance therapy guidelines developed by a joint expert panel consisting members of American Society for Blood and Marrow Transplantation (ASBMT), Center for International Blood and Marrow Transplant Research (CIBMTR), and EBMT [10, 11, 36]. ESMO guidelines recommend consolidation with brentuximab vedotin following HDC-ASCT in patients presenting with ≥ 1 of the following risk factors: primary disease progression, early disease recurrence after the end of frontline treatment (< 12 months), and extranodal disease at the time of relapse, without specifying the duration of consolidation therapy [11]. NCCN recommend brentuximab vedotin for 1 year for patients with a high risk of relapse, defined as ≥ 2 of the following risk factors: remission lasting < 1 year, extranodal involvement, PET-positive response at time of transplant, B symptoms, and/or > 1 salvage/subsequent therapy regimen [10]. The ASBMT, CIBMTR, and EBMT panels recommend post-ASCT consolidation with brentuximab vedotin for a maximum of 16 cycles every 3 weeks, or until unacceptable toxicity or disease relapse (whichever occurs first) for brentuximab vedotin-naïve patients with cHL and one or more high-risk features as defined by the AETHERA study [36]. The panel do not recommend brentuximab vedotin post-ASCT consolidation in patients whose HL has shown signs of refractoriness to prior treatment with brentuximab vedotin; however, high-risk patients with relapsed disease and limited prior exposure to brentuximab vedotin (~ 4–6 cycles) prior to ASCT and no evidence of brentuximab-vedotin-refractory disease, were recommended for brentuximab vedotin consolidation.

Following demonstration of the efficacy of brentuximab vedotin as consolidation in post-ASCT HL treatment, clinical studies to assess the efficacy of other anti-tumor agents, such as chemotherapies and checkpoint inhibitors, have been initiated. Novel therapies including ADCs, bispecific antibodies, and chimeric antigen receptor-modified T (CAR-T) cells are under investigation in R/R HL and other lymphomas, and may be an option for post-ASCT consolidation in the future. These are discussed briefly below.

Demonstration of an OS benefit remains the ideal clinical endpoint in the oncological studies, representing the time from randomization until death. Salvage with HDC-ASCT is an effective approach that can be further improved by consolidation strategies in high-risk patients [65]; however, owing to the effectiveness of targeted therapies applied later in the disease course, it can take many years to accumulate enough survival events to demonstrate a statistically significant difference in OS between an experimental therapy and the placebo/control arm. In the AETHERA study, patients had lengthy post-ASCT PFS and OS, with some patients not relapsing during follow-up, i.e., potentially achieving a cure [65]. After a median follow-up of 5 years there were still insufficient OS events to draw any conclusions on the efficacy of brentuximab vedotin compared with placebo in terms of impact on OS [17], and this may be further complicated by the use of subsequent checkpoint inhibitors. Crucially, AETHERA permitted patients in the placebo arm to crossover to the brentuximab arm, resulting in 87% of patients in the placebo arm eventually receiving brentuximab vedotin monotherapy as a subsequent therapy, further confounding assessment of the impact of brentuximab vedotin on OS [17, 65].

Based on the challenges faced when using OS as an endpoint, the European Medicines Agency recommended that time to second subsequent therapy (TTSST) is used as a measure of ongoing disease control and an alternative to OS (Fig. 1). TTSST is defined as the time between treatment initiation and the start of the third line of treatment (second subsequent therapy) and can be used to assess continuous lymphoma control in patient populations at high risk of relapse, in which multiple additional therapies are often used [86, 87].

In a post hoc analysis of the AETHERA study at 5 years, significantly fewer patients in the brentuximab vedotin arm had a TTSST event compared with the placebo arm (36% vs 46%, p < 0.0001), implying that 64% of brentuximab vedotin-treated patients had not received ≥ 2 subsequent therapies for HL or died, versus 54% in the placebo arm. Furthermore, subsequent transplants, including allo-SCT, were more common in the placebo arm than the brentuximab vedotin arm [17, 65].