Educational theme | The role of sequencing for MRD assessment and predicting relapse for ALL

The focus of our educational theme on the ALL hub this quarter is the role of sequencing in acute lymphoblastic leukemia (ALL) disease management. This article will explore the role of next-generation sequencing (NGS) in assessing minimal residual disease (MRD) and how this can help predict relapse in patients with ALL after tisagenlecleucel therapy.

A recent study by Pulsipher, et al., investigated the use of NGS with or without B-cell aplasia as a reliable biomarker to detect MRD and predict risk of relapse after tisagenlecleucel treatment in patients with ALL.¹ Treatment with tisagenlecleucel, an autologous, targeted immunocellular therapy, has shown long-term remissions in up to 50% of children and young adults with relapsed or refractory ALL, a group who were previously rarely curable. Despite the high percentage of remission, a portion of patients will lose their chimeric antigen receptor (CAR) T-cells within a few months. Since B-cell aplasia is a marker of CAR T-cell persistence, B-cell recovery or loss of B-cell aplasia suggests loss of functional CD19 CAR-T cells. In other patients, relapse may occur due to CD19 escape. Currently, no reliable markers exist to predict relapse in patients with ALL treated with CAR T-cell therapy. Detection of MRD by NGS at various time points after achieving remission with tisagenlecleucel could potentially stratify patients at high or low risk of relapse.

NGS-MRD detection and prediction of relapse risk after tisagenlecleucel therapy for ALL

The study assessed MRD detection and B-cell aplasia after tisagenlecleucel therapy for ALL to define biomarkers predictive of relapse in 143 patients. Blood and marrow samples from patients enrolled in the ELIANA (NCT02435849) and ENSIGN (NCT02228096) trials were analyzed and compared using multiparameter flow cytometry MRD (MFC-MRD) or NGS-MRD. Median follow-up in the two clinical studies was 38.4 months. Presence or absence of B-cell aplasia in the enrolled patients was considered as an indicator to evaluate patient outcome prediction.

There were no significant differences between the subgroup of patients with available NGS-MRD samples compared with those without (non-NGS), except for a trend towards fewer patients with prior hematopoietic cell transplantation in the non-NGS group (Table 1).

Table 1. Baseline patient characteristics in the NGS and non-NGS groups*

Sensitivity of MFC-MRD and NGS-MRD in bone marrow (BM) and peripheral blood

• MFC-MRD analysis detected a higher percentage of blasts in BM compared with peripheral blood. This was also observed in NGS samples, in which BMNGS-MRD detection was approximately one log higher.
• BMNGS-MRD resulted in comparable detection to BMMFC-MRD at levels above 10⁻⁴, but as expected, NGS-MRD was much more sensitive, showing high numbers of samples that were NGS-MRD positive (MRD⁺) but MFC-MRD negative (MRD⁻).
• Peripheral blood NGS-MRD was more sensitive at detecting disease compared to BMMFC-MRD.
• MFC-MRD limit of sensitivity was 10^-4, whereas NGS-MRD provides reliable sensitivity of ≥10⁻⁶.
• There were no samples that were MFC-MRD⁺ and NGS-MRD⁻.
• In many cases, disease was detected by NGS-MRD and not by flow cytometry.

MRD detection and time to relapse in the context of NGS-MRD detection quality and quantitation

• When analyzing NGS-MRD data, a number of patients had detectable disease but they were not reported as positive for the 10⁻⁶ cutoff because values did not reach thresholds required for the limit of detection or limit of quantitation.
• The more sensitive NGS measurements detect disease at levels that offer sufficient lead time prior to overt relapse to allow for repeat sampling and/or coordination of therapy options.

Predictive power of BMMFC-MRD and BMNGS-MRD

• Patients who achieved complete remission (CR)/CR with incomplete hematologic recovery (CRi) but did not achieve BMMFC-negative remission at Day 28 and/or Month 3 after tisagenlecleucel infusion did poorly.
  • 2/109 patients in CR/CRi at Day 28 were MRD-positive; these patients relapsed on Day 80 and 85 and died without additional therapy.

• BMNGS-MRD >0 in BM (at any sensitivity) was associated with higher risk of relapse compared to NGS-MRD =0 (NGS-MRD =0; 2-year event-free survival [EFS], 68%; 95% confidence interval [CI], 54–86 vs NGS-MRD >0; 2-year EFS, 23%; 95% CI, 8.8–62; p = 0.00047). Lower relapse rate led to higher overall survival (OS) in this cohort (2-year OS, 84%; 95% CI, 74–96 vs 2-year OS, 47%; 95% CI, 29–77 for NGS MRD =0 vs NGS-MRD >0, respectively; p = 0.0038).
• At 3 and 6 months after infusion, using positive BMNGS-MRD with a cutoff of 10⁻⁶ was predictive of poor outcome, but the best predictor at all time points was using a cutoff of NGS-MRD >0.

Predictive power of the loss of B-cell aplasia

• Kaplan-Meyer analyses of EFS in patients with continued B-cell aplasia vs patients who recovered B cells showed significantly worse outcomes in patients losing B-cell aplasia prior to Month 9 after tisagenlecleucel infusion.
• B-cell recovery within the first year of treatment was associated with significantly higher risk of relapse compared to patients without B-cell recovery (hazard ratio [HR], 4.5; 95% CI, 2.03–9.97; p < 0.001).

Multivariate and combined analysis models

• Multivariate analysis of time-dependent B-cell aplasia showed that at Day 28 after infusion CR/CRi patients with BMNGS-MRD >0 had significantly higher risk of relapse compared with those with BMNGS-MRD = 0 (HR, 4.87; 95% CI, 2.18–10.8; p < 0.001).
• At 3 months, in patients with BMNGS-MRD >0, HR increased to 12 (95% CI, 2.87–50; p < 0.001), whereas B-cell recovery was not independently predictive of subsequent outcomes (HR, 1.27; 95% CI, 0.33–4.79; p = 0.7).

Clinical associations with CD19⁺ or CD19⁻ relapse

• Relapses occurring with persistence of B-cell aplasia were largely CD19⁻ (88%).
• There was significantly more Grade III/IV cytokine release syndrome in patients with CD19⁻ relapse (64%; 16/25) compared with those with CD19⁺ relapse (22%; 3/14; p = 0.02). Higher CAR T-cell expansion in patients leads to higher grades of cytokine release syndrome.

Conclusion

In conclusion, the data obtained by combined analysis of the ELIANA and ENSIGN trials showed that BMNGS-MRD is currently the most sensitive biomarker to determine risk of relapse after CAR T-cell therapy in patients with ALL. Detectable BMNGS-MRD with or without B-cell aplasia can reliably predict relapse after tisagenlecleucel therapy with adequate time to consider approaches to relapse prevention such as hematopoietic cell transplantation or a second CAR T-cell infusion.