Effect of Brentuximab Vedotin Addition to Chemotherapy and Prognostic Factors in Patients with Relapsed/Refractory Hodgkin Lymphoma: A Large Multi-Trial Analysis Based on Individual Patient Data

Background

Brentuximab vedotin (BV) has been investigated in several single arm phase II trials as a new therapeutic option for patients with a first relapse or primary refractory classical Hodgkin lymphoma (R/R cHL), but no randomized controlled trials have been done that compare the addition of BV to salvage chemotherapy in the R/R setting. The aim of this study was to investigate the effect of BV addition to salvage chemotherapy compared to chemotherapy alone on progression free survival (PFS), overall survival (OS) and complete metabolic response (CMR) rate prior to autologous stem-cell transplant (ASCT), and to identify prognostic factors for response and survival.

Methods

We collected individual patient data of 718 transplant eligible R/R cHL patients treated in prospective clinical trials, of which 391 patients were treated with BV and salvage chemotherapy (BV-cohort), and 327 with salvage chemotherapy alone (Chemo-cohort), followed by ASCT [Table 1]. For PFS and OS analysis, the BV- and Chemo-cohorts were matched by propensity scores on baseline characteristics (i.e. relapsed/refractory, bulky disease, extranodal disease, stage I-II/III-IV, B-symptoms and first-line treatment with BEACOPP). Primary refractory disease was defined as not having achieved a CMR on first-line treatment. A CMR was defined as Deauville score (DS)1-3. PET-scans in the Chemo-cohort were assessed using IWG criteria (Cheson 2007) at the time of the original study. Therefore, we re-evaluated PET-positive scans using DS.

Results

After matching by propensity scores, there were no statistically significant differences in baseline characteristics between the BV- and Chemo-cohorts [Table 2]. The 3-year PFS was 74% (95%CI: 68-80%) in the BV-cohort compared to 67% (61-74%) in the Chemo-cohort (p=0.13) [Fig1A]. The 3-year OS was 94% (90-97%) versus 80% (74-86%); p<0.001, for BV- and Chemo-cohorts, respectively [Fig1B]. Patients with primary refractory disease did not show any difference in PFS between the BV- and Chemo-cohorts (p=0.50), while patients with relapsed disease showed a significantly higher PFS in the BV-cohort compared to the Chemo-cohort (p=0.023) [Fig1C+D].

When corrected for baseline characteristics, there were no significant differences in PFS between studies within the BV-cohort that used a sequential approach, multiple chemotherapeutic agents (e.g. DHAP/ICE/ESHAP vs bendamustine/gemcitabine) or different cumulative BV doses. Patients treated with a sequential approach with BV-ICE or ICE-GVD who achieved a CMR pre-ASCT on BV or ICE alone showed no difference in PFS compared to patients who achieved a CMR after BV-ICE or ICE-GVD (p=0.97).

Pre-ASCT PET response data were available for all patients in the BV-cohort and n=93 patients in the Chemo-cohort from the sequential ICE-GVD study. Patients with a CMR (n=349) had significant better PFS compared to patients with a partial response (n=67), with a 3-year PFS of 79% (75-84%) versus 58% (47-71%); p<0.001, respectively. Logistic regression (LR) corrected for baseline characteristics showed a significantly higher CMR rate for the BV-cohort compared to the CMR rate after ICE only (73% versus 61%; odds ratio (OR) 2.55; p<0.001), but there was no significant difference when compared to ICE-GVD (73% versus 84%; OR 0.73; p=0.34). B-symptoms (OR 0.52; p=0.02) and primary refractory disease (OR 0.43; p<0.001) were strongly associated with lower pre-ASCT CMR rates in a multivariable LR. Multivariable Cox regression for PFS showed a high prognostic value for pre-ASCT CMR (HR 0.42; p=0.004), and prognostic factors independent of pre-ASCT response were stage III/IV (HR 2.2; p=0.02), B-symptoms (HR 2.1; p=0.01) and primary refractory disease (HR 2.7 p=0.001).

Conclusions

The addition of BV to salvage chemotherapy followed by ASCT seems to increase PFS in relapsed, but not in primary refractory cHL patients. This suggests that in patients who are chemotherapy resistant other treatment modalities such as checkpoint inhibitors (CPI) should be considered. The observed increase in OS for the BV-cohort could be due to advances in treatment over time since novel therapies for patients who fail ASCT, such as CPI and BV, were not yet available at time of studies in the Chemo-cohort. Our study confirms the strong prognostic value of pre-ASCT CMR for PFS. B-symptoms, stage and primary refractory disease are prognostic factors for PFS and for achieving a pre-ASCT CMR.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal