CAR T versus ASCT for R/R B-cell NHL

This month, the Lymphoma Hub is focusing on the educational topic of chimeric antigen receptor (CAR) T-cell therapy. CAR T has shown promising efficacy in several settings, including relapsed/refractory (R/R) non-Hodgkin lymphoma (NHL). Currently, there are two CAR T-cell therapies - tisagenlecleucel and axicabtagene ciloleucel - which have been approved by the FDA and EMA for the treatment of NHL, specifically R/R diffuse large B-cell lymphoma (DLBCL).^1-4

The current standard of care treatment for patients with R/R B-cell NHL includes salvage autologous stem cell transplantation (ASCT). However, there are implementation challenges to using this approach universally.¹ These include: harvesting stem cells, the requirement of response to preconditioning chemotherapy, and patient ineligibility due to failure to mobilize stem cells or other severe complications.⁵

Some of these limitations can be addressed by CAR T therapy which has potential benefits over ASCT. CAR T therapy uses abundant, and easily-collectible, peripheral blood mononuclear cells (PMBCs), does not require a response to preconditioning chemotherapy and has the potential to be more clinically feasible as it may be possible to perform in a normal hematological ward or the outpatient setting.⁵ However, CAR T also comes with a unique side-effect profile, including cytokine release syndrome (CRS) and neurotoxicity, along with increased cost, and, due to its nature as a personalized therapy, the potential for manufacturing error.

Whilst there are similarities between ASCT and CAR T therapy, such as a common aim of reconstituting host immunological surveillance and inducing long-term remissions, no trial has yet prospectively addressed whether CAR T can provide a tangible benefit over salvage ASCT. Therefore, Caixia Li, the First Affiliated Hospital of Soochow University, Suzhou, CN, and colleagues, conducted a study to compare the outcomes of patients treated with CAR T therapy, to contemporaneous patients who received ASCT. This study was carried out at the First Affiliated Hospital of Soochow University.⁵

Study design and patient characteristics⁵

This study was conducted in patients receiving CAR T therapy (n= 29), whose outcomes were compared to contemporaneous patients who received an ASCT (n= 27). Baseline characteristics were similar, although patients in the CAR T group were slightly older, had higher International Prognostic Index (IPI) scores, a poor prognosis due to more prior lines of treatments, and a more advanced disease stage (Table 1).

Disease assessment immediately before treatment:

Given as CAR T versus ASCT unless otherwise stated

• Stable disease or progressive disease (PD): 82.8% vs 48.1%
• Remission: 17.2% vs 51.9%

Treatment-specific details⁵

CAR T therapy (anti-CD19)

Firstly, patients underwent autologous leukapheresis to obtain PMBCs for ex vivo CAR T manufacture. The subsequent conditioning chemotherapy given was fludarabine (30mg/m²) and cyclophosphamide (300mg/m²) on days -5, -4, -3. Patients then received an infusion of anti-CD19 CAR T-cells in doses from 5–10 x10⁶ CAR T-cells/kg.

ASCT

The source for ASCT was hematopoietic progenitor cells from the autologous peripheral blood. Stem cell collection was conducted with disease-specific chemotherapy and granulocyte colony-stimulating factor (G-CSF). The conditioning regimen was carmustine, etoposide, cytarabine and melphalan (BEAM) and busulfan (BU) / cyclophosphamide (CY).

• Number of CD34⁺ cells collected: 2.2–7.9 x10⁶/kg (median: 2.9 x10⁶/kg)
• Median time to from ASCT to neutrophil engraftment (ANC ≥ 0.5 × 10⁹/L): 10 days (8–15)
• Median time to from ASCT to platelet engraftment (PLT ≥ 20 × 10⁹/L): 12 days (9–25)

Efficacy⁵

Given as CAR T versus ASCT unless otherwise stated

The efficacy results are shown in Table 2 in the patients who were evaluable (25 vs 24). The remaining patients (4 vs 3) died prior to primary endpoint or were lost to follow-up. The median follow-up was 5 months for the whole cohort (5.2 months vs 4.7 months).

More patients in the CAR T group achieved a complete response (CR) compared to the ASCT group, and patients receiving CAR T-cell therapy also had a longer overall survival (OS) than patients undergoing ASCT. A subgroup analysis in patients with IPI scores ≥ 3 showed a significant benefit of CAR T in relation to ORR, CR, OS and PFS compared to ASCT.

Of the 12 patients achieving CR with CAR T, all 12 remained in CR at the latest follow-up. Of the six patients who achieved a partial response (PR), two remained in PR at the latest follow-up and four experienced PD in a median time of 5.3 months. Of the five patients achieving a CR with ASCT, four remained in CR post-ASCT, and one patient succumbed to multiple organ dysfunction syndrome.

Safety⁵

The safety data for the study are shown in Table 3. Overall, there were fewer non-hematologic severe adverse events (AEs, SAEs) in the CAR T group compared to the ASCT group (p= 0.03). There was a higher incidence of infection in the ASCT group compared to the CAR T group (p= 0.023), of which pulmonary infections were the most common. .

CAR T-related AEs

• CRS:
  • CRS occurred in 79.3% of patients, with 20.7% being grade ≥ 3
  • The most common AEs related to CRS were pyrexia (20.7%), hypotension (13.8%) and hypoxia (10.3%)
  • Median time from first infusion to CRS: three days (1–20)
  • Median time to resolution: four days
  • One patient died from severe CRS
  • Seven patients received tocilizumab and three receive glucocorticoids
• Neurotoxicity:
  • Neurologic events occurred in three patients, all of which were grade three or higher
  • The most common neurotoxicity events were confusion (10.3%) and aphasia (6.8%)
  • Median time from first infusion to neurotoxicity was 12.5 days (9–19).

Multivariate analysis⁵

In a multivariate analysis, the only factor significantly associated with PFS was an elevated lactate dehydrogenase (LDH) level, 95% CI, 0.085–0.732, p= 0.012. The only factors associated with OS were elevated LDH levels, 95% CI, 0.048–0.578, p= 0.005 (unfavorable) and CAR T-cell therapy, 95% CI, 0.090–0.641, p= 0.004 (favorable). Patient baseline characteristics, prior lines of therapy and disease status were not associated with OS or CR rates.

Limitations of the study⁵

• This study was conducted across different B-cell NHLs (predominantly DLBCL)
• The study did not assess tumor heterogeneity, the microenvironment and other factors impacting efficacy
• The outcomes of CAR T patients were compared to contemporaneous ones receiving ASCT as standard therapy which introduces a bias

Conclusion⁵

This study has shown that CAR T-cell therapy can provide superior outcomes compared to ASCT in relation to both safety and efficacy. However, the study is limited as it used a contemporaneous cohort of patients who received ASCT as the comparator.

The authors of the study suggest that the improved CR and OS of CAR T compared to ASCT indicate that CAR T may be a better therapeutic option in some patients with R/R B-cell NHL. This benefit was extended to patients with IPI scores equal to three or higher and noted that IPI scores were an independent unfavorable factor for OS and PFS in ASCT but not CAR T.

Two-cohort randomized controlled trials that directly compare CAR T to ASCT are required to determine the potential benefit of CAR T compared to ASCT in R/R NHL.