CIBMTR Analysis Confirms Ongoing Role for Autologous Transplant in Chemosensitive Relapsed DLBCL

Short abstract

In the era of chimeric antigen receptor (CAR) T‐cell therapy for relapsed/refractory diffuse large B‐cell lymphoma (DLBCL), a new analysis from the Center for International Blood & Marrow Transplant Research (CIBMTR) confirms an ongoing role for autologous transplant in patients with chemosensitive relapsed DLBCL.

For patients with relapsed, aggressive, chemotherapy‐sensitive diffuse large B‐cell lymphoma (DLBCL), high‐dose therapy with autologous hematopoietic stem cell transplant (auto‐HCT) has been an established consolidation procedure for 25 years. In 1995, the first prospective randomized trial of auto‐HCT versus conventional salvage chemotherapy in patients with relapsed aggressive non‐Hodgkin lymphoma showed a significant benefit in 5‐year event‐free survival (46% vs. 12%; p = .001) and 5‐year overall survival (OS) (53% vs. 32%; p = .038) in favor of transplant [1].

In 2014, the Center for International Blood & Marrow Transplant Research (CIBMTR) evaluated the role of auto‐HCT in patients with relapsed chemosensitive DLBCL based on the timing of relapse [2]. In the CIBMTR registry analysis, patients were stratified into two groups: early rituximab failure, with relapse ≤1 year from diagnosis (n = 297), and late rituximab failure, with relapse >1 year from diagnosis (n = 214). The clinical benefit of auto‐HCT was comparable across subgroups, with 3‐year progression‐free survival (PFS) rates of 44% and 52% in the early and late groups, respectively (p = .085). These findings confirmed auto‐HCT as a standard of care for chemosensitive relapsed DLBCL, regardless of the timing of relapse.

The introduction of chimeric antigen receptor (CAR) T‐cell therapy and approval of axicabtagene ciloleucel (axi‐cel) and tisagenlecleucel (tisa‐cel) in 2017 and 2018, respectively, changed the treatment landscape for relapsed/refractory DLBCL [3, 4]. Studies of axi‐cel and tisa‐cel demonstrated 2‐year PFS rates of approximately 30% to 40% in patients with ≥2 prior lines of therapy, establishing a new treatment approach for patients with heavily pretreated, relapsed/refractory DLBCL [3, 4].

According to CIBMTR registry data, auto‐HCT volume for DLBCL decreased by 45% between 2017 and 2018. This trend suggests that some patients with chemosensitive disease are proceeding to CAR T‐cell therapy rather than auto‐HCT. The current study was designed to test the hypothesis that auto‐HCT provides durable disease control for patients with DLBCL who have chemosensitive but positron emission tomography/computed tomography (PET/CT)‐positive disease at the time of transplant [5].

CIBMTR Analysis: Study Design

In the current study, investigators identified 1,938 patients in the CIBMTR registry who received their first auto‐HCT for DLBCL between 2003 and 2018. This cohort was narrowed to those patients (n = 249) with chemosensitive disease who received frontline rituximab‐based chemotherapy and achieved a partial response (PR) with PET/CT‐positive disease prior to transplant.

Study endpoints included OS, PFS, relapse, and non‐relapse mortality (NRM) in this patient population. Patients were stratified into two groups based on the timing of relapse: early chemotherapy failure, with relapse <12 months from diagnosis (n = 182), and late chemotherapy failure, with relapse ≥12 months following diagnosis (n = 67).

Patients who experienced early chemotherapy failure were significantly younger than those who experienced late failure (median age, 57 years versus 63 years; p < .01). In addition, patients with early failure were significantly more likely to have stage III–IV disease at diagnosis (74.2% vs. 53.7%; p = .003) and to have primary refractory disease after first‐line therapy (79.1% vs. 0%; p < .001).

CIBMTR Analysis: Key Findings

Lead study author Nirav N. Shah, M.D., M.S., of the Medical College of Wisconsin, presented findings on behalf of the CIBMTR. The 5‐year PFS was 41% for patients with relapsed chemosensitive DLBCL with PET/CT‐positive PR at the time of auto‐HCT, regardless of the timing of chemotherapy failure.

Results of the subgroup comparison showed no difference in NRM or relapse between the two groups (Table 1). The cumulative 5‐year incidence of relapse was 48% among patients with early failure and 57% for those with late failure.

Table 1.

Outcomes following auto‐HCT in relapsed chemosensitive DLBCL

Outcome	Early chemotherapy failure (n = 182)	Late chemotherapy failure (n = 67)	p value
Non‐relapse mortality
1‐year	7%	3%	.21
5‐year	10%	8%	.62
Relapse/progression
1‐year	41%	35%	.36
5‐year	48%	57%	.27
Progression‐free survival
1‐year	51%	66%	.03
5‐year	41%	41%	.93
Overall survival
1‐year	66%	85%	<.001
5‐year	51%	63%	.09

Abbreviations: auto‐HCT, autologous hematopoietic stem cell transplant; DLBCL, diffuse large B‐cell lymphoma.

Although 1‐year PFS was worse for patients with early failure relative to the late‐failure group (51% vs. 66%, respectively; p = .03), this difference disappeared over time. The 5‐year PFS was 41% in both groups (p = .93).

Survival outcomes showed a similar pattern. The 1‐year OS rate was worse in the early relapse group compared with the late relapse group (66% vs. 85%; p < .001). After 5 years, the difference was no longer statistically significant (51% vs. 63%; p = .09).

In the multivariate analysis, early chemotherapy failure significantly predicted worse OS relative to late chemotherapy failure (hazard ratio, 1.61; 95% confidence interval, 1.05–2.46; p = .03). By comparison, there was no association between timing of chemotherapy failure and relapse, PFS, or NRM.

According to the study authors, findings from this analysis support the ongoing use of auto‐HCT as a consolidative procedure for patients with relapsed chemosensitive DLBCL, even for those patients having PET/CT‐positive PR at the time of transplant. Moreover, although CAR T‐cell therapy is now an established standard of care for patients with chemorefractory DLBCL, more studies are needed to understand the optimal treatment approach for patients in early chemotherapy failure with chemosensitive DLBCL.