Response Assessment and Post–CAR T-Cell Therapy Management

Background

CD19 chimeric antigen receptor-T (CAR T) cell therapy has dramatically improved the outcomes for pediatric and adult patients with short-term complete remission (CR) rates of approximately 90% in relapsed/refractory B-lineage acute lymphoblastic leukemia (ALL) and objective response rates (ORRs) of 50% in non-Hodgkin lymphoma (NHL). Long-term, sustained remissions lasting ≥1 year after CD19 CAR T-cell therapy were demonstrated in 50% of patients with ALL and NHL. CD19 CAR T-cell therapy is associated with unique side effects that require specialized short- and long-term monitoring programs. A multidisciplinary approach with expertise in the treatment of patients with CAR T-cell therapy is needed with the aim of providing systematic, comprehensive follow-up for anticipated toxicities, disease monitoring, and CAR T-cell persistence.

Initial disease response assessments following CD19 CAR T cell therapy in patients with ALL are performed within the first month, standardly between days 21 and 28 postinfusion. The testing for ALL includes bone marrow (BM) aspirate evaluation for morphology and minimal residual disease (MRD) assessment, either by multiparameter flow cytometry (MFC), polymerase chain reaction (PCR), or immunoglobulin-based next-generation sequencing (Ig-NGS). In addition, cerebrospinal fluid (CSF) is evaluated for evidence of malignant disease, and other extramedullary sites, if present, are evaluated by positron emission tomography (PET) imaging. Initial BM disease responses are characterized as either CR or CR with incomplete hematologic recovery (CRi or CRh; platelet count <100,000/μL or absolute neutrophil count [ANC] <1,000 cells/μL) and either MRD-negative (<0.01% of mononuclear cells [MNCs] by MFC) or MRD-positive (≥0.01% of MNCs by MFC) CR. CAR T-cells are detected either via flow cytometry using antibodies for constructs producing surface markers or quantitative PCR (qPCR) assays for CAR transgene levels. However such assays are not routinely performed for the two FDA-approved CAR products. A surrogate marker of CD19 CAR T-cell persistence is B-cell aplasia (BCA), which is monitored via flow cytometry as absolute CD19 counts in the peripheral blood.

In NHL, disease responses are monitored via whole-body imaging techniques, including computed tomography (CT) and fluorodeoxyglucose (FDG)-positron emission tomography (PET) imaging before and after CAR T-cell therapy. Initial disease assessments in NHL are reported according to the International Working Group Response Criteria for Malignant Lymphoma (Lugano criteria) as ORR, which include CR (which also includes a complete metabolic response even with a persistent mass) and partial response (PR; a decrease by >50% in the sum of the product of the perpendicular diameters of up to six representative nodes or lesions). Importantly, NHL patients with PR at initial disease assessment, usually 1–2 months post–CAR T-cell therapy, can continue to respond and convert to CR in >50% of cases. Stable disease (SD) or progressive disease (PD) are other disease response categories commonly employed in the setting of NHL.

After typically inpatient care of acute toxicities, including cytokine release syndrome (CRS), neurotoxicity (reviewed in Chapters This needs to be clarified based on the most up to date chapter index), and neutropenic fever/infection, subsequent management is focused on correcting and supporting any remaining complications associated with CAR T-cell therapy in an outpatient setting. This includes management of cytopenias, which may occur in up to 30%–60% of patients, with packed red blood cell (PRBC) and platelet transfusions, growth factor support with eltrombopag for persistent severe thrombocytopenia, and granulocyte colony-stimulating factor (G-CSF, filgrastim), for severe neutropenia. G-CSF administration can be considered, either preemptively, at the initial onset of neutropenia, or later, after resolution of CRS, without firm data available regarding the optimal timing of administration. In addition, patients require monthly immunoglobulin (intravenous or subcutaneous) infusion for hypogammaglobulinemia as a result of BCA to maintain adequate IgG levels (>400 mg/dL). The long-term complications after CAR T-cell therapies have only recently begun to be systematically evaluated, given the relatively recent incorporation of CAR T-cells into the treatment of B-cell malignancies. These complications include prolonged cytopenias, infections, and specific organ dysfunction, such as cardiac and neurologic dysfunction as well as renal insufficiency that can persist after resolution of CRS and neurotoxicity. Such specific complications dictate the need for personalized long-term follow-up, which will become increasingly relevant as the utilization of CAR T-cell therapy increases and as more patients enter the 15-year follow-up period mandated by the FDA for recipients of gene-modified cellular therapy.

CD19 CAR T-Cell Efficacy in Relapsed/Refractory Pediatric and Adult ALL ( Tables 9.1 and 9.2 )

CD19 CAR T-cell therapy induces CR in 70%–93% of adult and pediatric patients with relapsed and refractory ALL. Almost all of the morphologic CRs reported within 1 month post–CAR T-cell infusion are MRD-negative by MFC. Published studies of CARs containing the CD28 costimulation domain report CR rates of 70% in pediatric patients and 80%–83% in adults, with MRD-negative CRs in 60% of children and 67%–80% of adults. Clinical trials of 4-1BB-CARs report CR rates of 81%–93% in pediatric patients and 82%–97% in adults, with MRD-negative CRs in 73%–93% of children and 78.5%–90% of adults. The durability of response is CD19 CAR construct dependent, with more durable responses reported with 4-1BB-CARs compared with CD28-CARs ( Tables 9.1 and 9.2 ). In adult studies, reported event-free survival (EFS) rates range between 50% (at 8 months) and 70.5% (at 300 days) with 4-1BB-CARs versus 39%–50% (at 6 months) with CD28-CARs. In pediatric studies, 1-year EFS rates of 50% have been observed with overall survival (OS) rates of 66%–76% following 4-1BB-CAR T-cell therapy, with reported leukemia-free survival (LFS) of 78.8% (at 4.8 months) and OS of 51.6% (at 9.7 months) following CD28-CARs. This important difference in the durability of response between CD28-CARs and 4-1BB-CARs has resulted in more frequent use of hematopoietic stem cell transplantation (HCT) after achieving an MRD-negative CR with CD28-CARs. In pediatric studies, 83% of patients underwent HCT following CD28-CARs compared with 10%–26% of patients after treatment with 4-1BB-CARs. In addition, an apparent survival benefit of consolidative HCT has been observed in children who achieve an MRD-negative remission after CD19 4-1BB CAR T-cell therapy in individuals who have not previously undergone HCT.

While CD19 CAR T-cell therapy results in high rates of MRD-negative remissions in relapsed/refractory ALL within the first month of therapy, the durability of these responses is less encouraging, with 1-year EFS of 50% in pediatric patients and <50% in adults ( Fig. 9.1A–C ). Studies exploring factors associated with continued persistence of CAR T-cells have defined multiple factors associated with durable responses: antigen burden, type of lymphodepletion chemotherapy, CAR construct type, early maximum CAR T-cell expansion, Ig-NGS MRD-negative status post–CAR T-cell therapy, and BCA lasting >6 months ( Table 9.3 ). In a pediatric study of CD19 4-1BB-CD3ζ CAR T-cells of defined composition, i.e. 1:1 CD4:CD8 ratio, CD19 antigen burden of >15% in the BM was associated with increased persistence of BCA ( Fig. 9.1E ). The same study also demonstrated improved BCA duration in patients who received a combination of fludarabine and cyclophosphamide lymphodepletion, which was also confirmed in an adult study ( Fig. 9.1D ). In a study of tisagenlecleucel, measurements of CAR transgene levels by qPCR revealed that earlier maximum expansion, at a median of 10 days, occurred in responding patients, compared with a delayed expansion, at a median of 20 days, in nonresponders. Ig-NGS methods have been implemented to monitor MRD in pediatric and young adult patients with ALL who have received tisagenlecleucel therapy and demonstrated superior MRD detection compared with MFC, with increased lead time to morphologic relapse of 67 versus 39 days. This approach using NGS-MRD testing allows earlier identification of patients at high risk for relapse and provides a window of opportunity for such patients to proceed rapidly to HCT before morphologic relapse. In addition, patients who were NGS-MRD-negative at day 28 following CD19 CAR T-cell infusion demonstrated improved duration of response with an 80% chance of remaining relapse-free at 3 years. It is of critical importance to be able to distinguish CAR T-cell –induced CRs that will persist without any additional therapy from those that are short lived and use this distinction to avoid the toxicities of HCT in patients who may not otherwise need additional therapy. As BCA is a pharmacodynamic surrogate marker of functional CAR T-cell persistence, early B-cell recovery within 6 months of tisagenlecleucel infusion was associated with early loss of CAR T-cells and potentially relapse ( Table 9.3 ).

Long-Term CD19 CAR T-Cell Responses in Relapsed/Refractory Pediatric and Adult ALL ( Tables 9.1 and 9.2 )

Long-term responses in patients with relapsed and refractory ALL have only recently been systematically evaluated, and very few large studies report responses >1 year duration given the relatively recent introduction of this therapy in the pediatric and adult setting. In pediatric studies, 1-year EFS of approximately 50% has been reported with CD19 4-1BB-CD3ζ CARs, while OS at 9.7 months was 51.6% with CD19 CD28-CD3ζ CARs. HCT following treatment with CD19 4-1BB-CD3ζ CAR T-cells was performed in 10%–26% of pediatric patients in CR and resulted in 90%–100% survival. Of 12 patients in CR following treatment with CD28-CD3ζ CAR T-cells, 10 patients (80%) received consolidative HCT, and 100%, all 10, were alive at the time of reporting.

Long-term responses in older adults with relapsed and refractory ALL show similarly durable responses. In 30 patients treated with CD19 4-1BB-CD3ζ CARs, disease-free survival (DFS) was 70.5% with 300 days of median follow-up. In a single institution study of adults with relapsed, refractory, or MRD + ALL treated with CD19 4-1BB-CD3ζ CARs, 8-month EFS was 50% for patients who did not receive HCT, while 133-day EFS was 85% for patients who received HCT. In the largest follow-up study to date, of 53 adults treated with CD19-CD28-CD3ζ CARs, median EFS was 6.1 months, and median OS was 12.9 months. In this single center study, patients with low disease burden (<5% BM blasts) had significantly improved EFS and OS compared with patients with high disease burden (≥5% BM blasts or extramedullary disease), with EFS of 10.5 versus 5.3 months and OS of 20.1 versus 12.4 months. These findings are in contrast to those of 4-1BB-CD3ζ CARs, where a high disease burden (>5% or >15% BM blasts) was associated with improved CAR T-cell persistence and outcomes. The authors of the CD28-CD3ζ CAR report speculated that patients with low disease burden were more likely to achieve a high CAR T-cell to leukemia cell ratio that is required for effective disease eradication. Although multiple prior studies had correlated longer CAR T-cell persistence with improved survival, this was not the case for the CD19-CD28-CD3ζ CAR T-cell study, in which the median CAR T-cell persistence of 14 days was not associated with longer survival. Studies have demonstrated T-cell–mediated rejection responses directed at murine single-chain variable fragment (scFv) CAR epitopes as contributing to limited CAR T-cell persistence, and humanized and fully human CD19 4-1BB-CD3ζ CARs have been developed. Thus far, published data from phase 1 clinical trials is promising using humanized CD19 4-1BB-CD3ζ CARs, with 6-month LFS of 65.8% and OS of 71.4%, results that are similar to those with murine CD19 CARs, though data on CAR persistence with the humanized construct is limited.

Limited CAR T-Cell Persistence: Therapeutic Options

For patients with limited CAR T-cell persistence, a second infusion of CAR T-cells has been performed, with or without repeat lymphodepleting chemotherapy using varying intensity (sometimes increased); however, there are limited reports of engraftment and antileukemic activity. The addition of checkpoint inhibitors, such as pembrolizumab, has been associated with CAR T-cell reexpansion, persistence, and objective responses in a subset of patients in early studies. In our center, we routinely monitor BCA in patients who have received CD19 CAR T-cell therapy. For patients who demonstrate loss of BCA (absolute peripheral blood CD19+ lymphocyte count >50 cells/μL), especially within the first 6 months following infusion, we recommend a second CD19 CAR T-cell infusion followed by a checkpoint inhibitor, such as pembrolizumab. If the patients continue to demonstrate poor CAR T-cell persistence and loss of BCA but remain in MRD-negative remission (by MFC), we recommend a consolidative HCT, especially for patients who have not previously undergone HCT.

Efficacy of CD19 CAR T-Cell Therapy in Adult NHL ( Table 9.4 )

High response rates after CD19 CAR T-cell therapy have also been demonstrated in NHL, especially in diffuse large B-cell lymphoma (DLBCL), follicular lymphoma (FL), and mantle cell lymphoma (MCL). Best reported ORRs, which include CRs and PRs, range between 52% and 100%. In an international, multicenter study of tisagenlecleucel in patients with DLBCL and transformed FL (TFL), 37 of 93 (40%) of patients achieved a CR at a median of 2 months (range 1–17 months), including 16 patients who had either SD (n = 4) or PR (n = 12) at 1 month after infusion. A conversion from PR to CR occurred in 13 of 24 (54%) of patients in the study. Importantly, remission at 3 months was predictive of long-term remission at 12 months, in which 28 of 35 (81%) of patients who were in remission by 3 months remained in remission at 12 months.

The pivotal multicenter, phase 2, ZUMA-1 trial examined the efficacy of axicabtagene ciloleucel (CD19 CD28-CD3ζ CAR T-cells) in 101 patients with relapsed or refractory DLBCL, primary mediastinal B-cell lymphoma (PMBCL), and TFL. Best ORR included CR (disappearance of measurable disease on CT scan or residual, PET negative masses) and PR (≥50% decrease in tumor burden with ongoing PET avidity) in 82% of patients, 54% of patients with CR, and 28% with PR, with a median response time of 1 month. Two-year follow-up results demonstrated an OS of 51% and PFS of 39% at a median follow-up of 27.1 months. In addition, the follow-up results confirmed that the responses were durable, in which 93% of patients who demonstrated a response at 12 months continued with response at 2 years.

The international, phase 2 study, JULIET, using tisagenlecleucel (CD19 4-1BB-CD3ζ CAR T-cells) in patients with relapsed or refractory DLBCL demonstrated ORRs and CR rates of 52% and 40%, respectively. Six-month ORRs and CR rates were 33% and 29%, respectively. Remissions were durable; 1-year relapse-free survival (RFS) was 65% and OS was 49%. Similar responses were also demonstrated in a multicenter phase 1 trial with lisocabtagene maraleucel (JCAR017; CD19 4-1BB-CD3ζ CAR T-cells) in patients with relapsed or refractory DLBCL: best ORRs and CR rates were 80% and 52%, respectively, and 3-month ORR was 74% in the core patient cohort that included DLBCL not otherwise specified (NOS) and high-grade lymphoma only. Durability of responses was similar to other CD19 CAR T-cell trials in NHL; 1-year RFS was 42%.

Long-Term CD19 CAR T-Cell Responses in Adult NHL ( Table 9.4 )

In NHL, long-term, durable CAR responses were demonstrated in three recent key trials: ZUMA-1, TRANSCEND NHL, and JULIET, with 1-year PFS of 44% and 1-year RFS of 42% and 65%, respectively. One-year OS was 59% in ZUMA-1 and 49% in JULIET. Importantly, some patients without CR at 1 month postinfusion converted to CR (11 of 35 with PR; 12 of 25 with SD) as late as 15 months after treatment, and those CRs were sustained with few relapses. Notably, there were no relapses observed in ZUMA-1 patients who were in CR at 1 year.

Post–CAR T-Cell Therapy Management

CD19 CAR T cell therapy is associated with unique toxicities and complications that require a variety of complicated management strategies in at least 50% of patients during the acute setting, and all patients require long-term follow-up monitoring after CAR T-cell therapy. Those who receive CAR T-cell infusion in the inpatient setting are monitored in the hospital for CRS and immune effector cell–associated neurotoxicity syndrome (ICANS) until resolution of clinical and biochemical markers of toxicities. Following discharge, patients are monitored closely in the outpatient setting. For patients who receive CAR T-cells as an outpatient procedure, follow-up visits are usually conducted two to three times per week for the first 3–4 weeks to monitor for CRS, ICANS, the need for transfusions, and symptoms of infection. Almost all patients will develop fever, at which point hospitalization for monitoring and care is needed until CRS is resolved. Generally, at day 28 post–CAR T-cell therapy, disease reevaluation is performed. Patients subsequently have at least monthly visits for IVIG supplementation to manage hypogammaglobulinemia, which is a direct consequence of CD19 CAR T-cell–mediated BCA. Consensus guidelines for the management of acute CD19 CAR T-cell therapy–associated toxicities and complications have been established for pediatric and adult patients. However, systematic, comprehensive long-term evaluation and management guidelines have not yet been developed and adopted by the CAR T-cell community. The recommendations below are a systems-based approach guided by a comprehensive review of published reports of CD19 CAR T-cell–mediated toxicities.