The rapidly evolving role of CAR T-cell therapy in the treatment of relapsed/refractory Hodgkin lymphoma

Hodgkin lymphoma (HL) is a B-cell malignancy characterized by malignant Hodgkin and Reed-Sternberg (HRS) cells within an inflammatory cellular infiltrate. An article summarizing the recently updated British Society for Haematology guidelines for first-line management of classical HL (cHL) can be found on the Lymphoma Hub. Although most patients are cured with frontline treatment, disease progression or relapse may occur in upwards of 20–30% of patients at some point in their lifetime. In these scenarios, standard care has focused on highly intensive chemotherapy followed by autologous hematopoietic stem cell transplantation (auto-HSCT). A further article on the Lymphoma Hub provides an overview of research that has incorporated novel therapies, such as brentuximab vedotin (BV), checkpoint inhibitors, and response-adapted treatment approaches, in patients with relapsed/refractory (R/R) cHL.

The recent emergence of targeted immunotherapies that reprogram the patient’s immune system to overcome the immunosuppressive tumor microenvironment (TME) and improve cancer cell clearance has been revolutionary. In particular, the development of chimeric antigen receptor (CAR) T-cell therapy, whereby the patient’s own immune cells are engineered ex vivo to recognize target cancer antigens, has shown promise in trials for non-Hodgkin lymphoma.

Chimeric antigen receptor (CAR) T-cell therapies are widely used to treat relapsed/refractory (R/R) B-cell malignancies, with four CAR T-cell products currently approved in Europe: tisagenlecleucel for pediatric B-cell acute lymphoblastic leukemia and adult large B-cell lymphoma, axicabtagene ciloleucel for adult large B-cell lymphoma, KTE-X19 for adult mantle cell lymphoma, and idecabtagene vicleucel for adult multiple myeloma. CAR T-cell therapy is associated with potentially life-threatening immunological toxicities. An article summarizing the management of short-, medium-, and long-term complications in patients receiving CAR T-cell therapy can be found on the Lymphoma Hub.

In November 2021, the Lymphoma Hub published an article summarizing a discussion of CAR T-cell therapy in patients with R/R HL by Ho, et al. Here, we build on our understanding of the emerging role of this therapy with additional information from Meier, Savoldo, and Grover,¹ as published in Journal of personalized medicine. In particular, the authors discuss recent data on strategies to optimize and determine the full potential of CAR T-cell therapy.

Predicting patient response to CAR T-cell therapy

Voorhees, et al., (NCT02690545) sought to develop a predictive model of responders to CD30 targeted CAR T-cells in a bid to better understand which patients may benefit most from this therapy. They assessed metabolic tumor volume (MTV) by measuring tumor burden using 18F-fluorodeoxyglucose positron emission tomography imaging based on the maximum standardized uptake value. A direct correlation between MTV ahead of CAR T-cell therapy and progression-free survival (PFS) was noted; patients with a high MTV prior to lymphodepletion and CAR T-cell infusion had a significantly lower 1-year PFS (14%) compared to those with low MTV (58%). Patients whose MTV was reduced following bridging therapy prior to CAR T-cell treatment also had improved 1-year PFS (40%) compared to those whose tumor burden remained high (1-year PFS 0%), although bridging therapy itself was not associated with a significant difference in PFS. The correlation between MTV and PFS suggests that this may be a useful parameter to risk stratify patients receiving CAR T-cell therapy, and that bridging therapy to lower MTV ahead of lymphodepletion may lead to improved outcomes.

Targeting the tumor microenvironment
Given the relatively sparse number of HRS cells admixed in a much larger population of non-malignant inflammatory cells, CAR T-cell therapy targeting the TME is under investigation. The TME plays a significant role in the immune evasion and disease progression seen in cHL, as discussed in more detail in our previous article on the hub. Recruitment of tumor-associated macrophages (TAMs) supports the immunosuppressive TME, and an increased percentage of TAMs has been linked with decreased overall survival in patients with HL.

A preclinical study of CAR T-cell therapy targeting CD123 (a molecule found on TAMs and malignant HRS cells) by Ruella, et al., demonstrated their ability to eradicate cHL cells in a mouse xenograft model and develop immunological memory such that when mice were re-challenged with HL cells, a significant re-expansion of CAR T-cells was seen and there was no tumor growth.

Another target to be considered is CD19, as a means of targeting non-malignant B cells in the TME as well as potentially CD19 positive HRS clones that are known to be in circulation. In a small cohort of heavily pretreated patients, CAR T-cell therapy well tolerated, and a 40% response rate was seen. Clearly more research is needed in terms of targeting the immunosuppressive cells of the TME such as TAMs, however it certainly seems a promising area of development.

Improving CAR T-cell persistence
Circulating CAR T-cells have been detected in peripheral blood up to 8 weeks post-infusion, though it is thought they may exist for longer in the TME and tissues. It has been proposed that improving the longevity of CAR T-cells will improve their cancer-killing ability. Most CAR T-cells are terminally differentiated effector cells due to the gene manipulation that is required to produce CAR expression. This leads to increased exhaustion and altered T-cell homing which reduces their efficacy in vivo. Less differentiated adoptively transferred T-cells have much higher anti-tumor potential, particularly a subset of memory stem T-cells (T_scm). CD30 targeted CAR T-cells enriched with T_scm-like cells were found to have superior anti-tumor activity in a mouse xenograft HL model compared to cells without. In addition, these T_scm-like cells were exhaustion resistant in the presence of repeated antigen stimulation and prevented tumor growth in a re-challenge model. Despite the potential therapeutic advantage of increasing T_scm cells, how best to generate these cells is unclear.

A potential means to increase these memory cells could be via the design of the CAR T-cell. When evaluating the effect of co-stimulatory domains on the cells’ phenotype, CAR T-cells that expressed the 4-1BB (CD137) were more likely to be central memory cells and persist longer than those that expressed CD28, which tended to be effector-like. Early limited clinical data showed a comparable duration of detectable CAR T-cells in peripheral blood using either 4-1BB or CD28. The effect of either co-stimulatory domain on CD30 targeted CAR T-cells is yet to be determined; however, optimizing these domains certainly seems to be beneficial, as shown in a recent study with xenograft mouse models in which CD30 CARs containing two co-stimulatory domains (CD28 and OX40) lived longer and had greater anti-tumor activity

Finally, the lymphodepleting agent fludarabine may also play a role in CAR memory T-cell development, with studies showing increased memory formation in both CD4 and CD8 T-cell compartments with its use. In a study by Ramos, et al., (NCT02690545 and NCT02917083), patients who had received fludarabine were found to have higher circulating levels of interleukin(IL)-7 and IL-15 and longer CAR T-cell persistence. Given that these cytokines have been linked to memory T-cell formation, further strategies that increase their production could also theoretically improve CAR T-cell persistence. For example, preclinical studies in which IL-7 gene expression was coupled with the CAR showed increased populations of memory CD8 T-cells.

Enhancing CAR T-cell trafficking

Another mechanism to enhance CAR T-cell efficacy is to improve their trafficking to the TME. In our previous article, we touched upon anti-CD30 CAR T-cells engineered to co-express CCR4, given this is the ligand for several chemokines secreted by HRS cells that aid immune evasion and inhibit immune response. Cytotoxic T-cells do not express CCR4 and are therefore not readily recruited to the TME. Preliminary results from a phase I dose-escalation trial (NCT03602157) included eight patients with HL, all of which responded to treatment with CD30 CAR T-cells co-expressing CCR4, with 75% having a complete response with median PFS not having been reached after a median follow-up of 12.7 months. Promisingly, significant enrichment of CAR T-cells was seen at the tumor site in comparison to peripheral blood, supporting the theory that further modification of CAR T-cells can improve efficacy and response durability.

Combined checkpoint inhibition and CAR T-cell therapy

A mechanism by which CD30 CAR T-cell efficacy may be impaired is via retained expression of programmed cell death protein 1, which engages with the upregulated programmed cell death protein 1 ligands to inhibit T-cell activation and proliferation. A study that evaluated checkpoint inhibitors in patients whose disease had progressed after CAR T-cell therapy observed clinical benefit, with all 5 patients responding. Interestingly, this included those who had not responded to checkpoint inhibition prior to CAR T-cell therapy. Meier, et al.,¹ speculate that some of the responses observed could be due to reprogramming and reactivation of memory CAR T-cells that had persisted after the initial infusion. A phase I trial (NCT04134325) is currently evaluating the efficacy of immunotherapy after CAR T-cell therapy failure. The possibility of an upfront combination of CAR T-cell therapy and checkpoint inhibition may be a more effective strategy and is currently being explored in solid tumors.

The interplay of transplant and CAR T-cell therapy

Whilst we have gained some clarity on the role of CAR T-cells in patients with R/R HL who are ineligible for auto-HSCT, or who have relapsed posttransplant, the question remains as to how CAR T-cell therapy can benefit those patients eligible for transplant. Auto-HSCT currently offers the highest curative potential in these patients. More novel agents such as BV and checkpoint inhibitors are less likely to offer long-term disease control but given the significant transplant-associated morbidity and the occasional durable remissions seen with these agents, they are often trialed first. Patients who have a complete response to salvage therapy prior to auto-HSCT and/or who have chemosensitive disease tend to have better long-term outcomes; therefore, given the success of CAR T-cell therapy even in heavily pretreated patients, the authors postulate a potential role for CAR T-cell therapy as a bridge to transplant, though this has yet to be explored in HL.

The role of CAR T-cell therapy as a maintenance or consolidation strategy posttransplant is currently being considered (NCT04083495). In this phase I trial, patients deemed high-risk received CAR T-cell therapy post-auto-HSCT. The 1-year PFS of 15 patients included was 79%, with an OS of 100%, results that the authors suggest are at least comparable to the AETHERA trial (NCT01100502) in which BV was used as a maintenance strategy in BV naive patients. With 80% of the patients in the NCT04083495 trial having seen BV at some point prior to transplant, CAR T-cell therapy may be particularly appropriate for patients who have previously received BV or are no longer able to receive it due to neuropathy.

Concern about the possibility of an increased incidence of immune-related adverse events post-, auto-, or allo-HSCT has largely been disproved by multiple studies demonstrating long-term safety of CAR T-cells.

It is still unclear whether CAR T-cell therapy will ever replace HSCT as the standard of care in R/R HL. An intent to treat analysis in patients with R/R non-Hodgkin lymphoma demonstrated CAR T-cells were not inferior to allo-HSCT. Whilst they carried the added benefit of reduced non-relapse mortality, patients who received CAR T-cells had a higher tendency for relapse. Clearly, a significant amount of further research is needed before we fully reform the standard care for transplant eligible patients.

Conclusion

The high response rates seen with CAR T-cell therapy with the potential for durable responses, as well as limited toxicity, is clearly very promising. Further work is needed to optimize this therapy and clearly determine how it fits in with other therapeutic strategies, including whether in time we will rethink the timing and role of transplantation. In this rapidly evolving landscape, the future of treatment, even for patients with the most refractory disease, looks promising.