Abstract: Enthusiasm for immunotherapy has spread throughout the oncology community in recent years, as therapies harnessing the body’s ability to fight cancers have offered the potential for remission and even cure in patients with otherwise few options. Chimeric antigen receptor–modified T cells (CAR-Ts) have emerged as a new entity in the treatment armamentarium for several hematologic malignancies, including non-Hodgkin lymphoma. CAR-T therapies are expected to begin receiving approvals from the US Food and Drug Administration as early as 2017. Although commercial pricing data for CAR-T therapies are not currently available, these treatment regimens are anticipated to carry a large financial burden, due to complex manufacturing requirements and the high associated toxicity profile. In this article, the authors review the foundational research in CAR-T therapy, including pricing estimates, in order to determine how best to integrate this new branch of immuno-oncology into clinical pathways for patients with B-cell malignancies.
Citation: Journal of Clinical Pathways. 2017;3(2):41-47.
Received January 26, 2017; accepted March 7, 2017
Practicing oncologists have generally used expert opinions, published guidelines, and consensus statements as the basis for making therapeutic decisions that balance efficacy with toxicity while factoring patient choices and comorbidities.1 This evidence-based decision-making process has led to the use of clinical pathways, governed by oncologists and/or payers, which are hypothesized to help providers deliver more efficient, more streamlined, and better care.2 The principles underlying clinical pathway development are that the most effective therapy be selected when available; when regimens are equally effective, the one with the most favorable toxicity profile is recommended; and, if efficacy and toxicity are comparable, the most cost-effective therapy is suggested.1
The past decade has witnessed significant improvements in the prognosis of patients with non-Hodgkin lymphoma (NHL), regardless of histology. Patients are now expected to experience better overall response rates (ORR), longer disease-free intervals, and improved overall survival.3 These improvements are multifactorial, owing to better therapeutics, monoclonal antibodies, supportive care, and personalized therapy.3 Despite these improvements, a significant proportion of patients will relapse or develop refractory disease.
Genetically engineered T cells, often referred to as “living drugs”, are emerging as a new modality in the fight against cancer.4 Enthusiasm for harnessing the immune system to treat malignancies has been reinforced since the recognition of the graft-versus-leukemia effect and the curative potential of allogeneic hematopoietic stem cell transplantation (allo-HSCT).5 The potential role of immunotherapy in treatment, augmenting and perhaps bypassing the role of traditional or targeted chemotherapy and/or radiation, may benefit patients with refractory or relapsed disease.
Innate and adaptive immunity are equally important host defense mechanisms against pathogens and malignancy. Main components of the innate immune system are epithelial barriers, phagocytic leukocytes, dendritic and natural killer cells, and circulating plasma proteins.6 Components of the adaptive immune system are usually silent but are activated when needed.7 The two subtypes of the adaptive system are humoral immunity, mediated by antibodies produced by B lymphocytes (B cells), and cell-mediated immunity, led by T lymphocytes (T cells).7 T cells identify cancer cells through antigen recognition, mediated by T-cell receptors (TCRs). These receptors are linked to peptides presented by the major histocompatibility complex; binding of the antigen complex to the receptor initiates T-cell activation and elicits an antitumor response.8 Graft-versus-host-disease (GVHD), one of the most hazardous complications of allo-HSCT, is the immune response of donor T-cells in reaction to the recipient’s alloantigens. Manipulation of donor T cells to prevent major histocompatibility complex antigen recognition has been commonly utilized as a strategy to reduce GVHD complications while maintaining graft effectiveness.9
Our understanding of the etiology of GVHD, and the resulting development of T-cell manipulation, has led to the hypothesis that autologous T cells could be manipulated to recognize malignant cells in a way that allows tumor control without the negative sequelae of GVHD.8 This requires that the T cells be harvested via apheresis and transported to a laboratory where they are chemically modified by linking the extracellular antigen recognition domain from a monoclonal antibody fragment to the T cell’s intracellular signaling domain.10 This newly modified autologous T cell–antigen receptor complex is a chimera of two proteins. The newly created chimeric antigen receptor–modified T cells (CAR-Ts) are then incubated to expand their number. Once adequately expanded, the CAR-Ts can be infused into the patient, but only after administering chemotherapy that depletes the patient’s own circulating lymphocytes to maximize the therapeutic effectiveness of the CAR-Ts.11
Hematologic malignancies that are defined by the expression of unique antigens on their cells have become the disease prototype for this form of targeted immunotherapy. Several CAR-T therapies are being investigated in clinical trials, and regulatory approval of one or more of these CAR-Ts is projected in 2017. Relapsed/refractory NHL is the most studied of all hematologic malignancies in clinical trials evaluating the role of CAR-T therapy. For this reason, early CAR-T therapy approval is most likely to be seen in this setting or in the pediatric acute lymphoblastic leukemia (ALL) setting.
Available data have demonstrated CAR-T therapy to have remarkable activity (Table 1), but with moderate to excessive toxicity. Additionally, costs are anticipated to be high. How this novel therapeutic modality will be adopted in a value-based care environment is therefore controversial. For CAR-T therapies to be incorporated into any clinical pathway, they will have to be value tested, which necessitates a thorough understanding of their efficacy, toxicity, and cost.
Other than stem cells, B cells ubiquitously express CD19 on their surfaces throughout their development; as such, this antigen has become an ideal target for CAR-Ts. Several studies of therapies targeting this receptor have been conducted in patients with relapsed and/or refractory CD19-positive disease. Original studies in chronic lymphocytic leukemia (CLL) have shown complete responses (CRs) in heavily pretreated patient populations, with some attaining minimal residual disease (MRD) negativity, a hallmark of efficacy.12-14 In relapsed/refractory ALL, where salvage therapies are considered a bridge to allo-HSCT,15 CAR-Ts showed rapid responses and MRD negativity.16 In fact, Maude et al17 treated 30 adult and children patients with escalating doses of CAR-Ts and found a CR in 27 patients (90%), including 15 who failed prior transplantation; MRD negativity was attained in 22 patients (73%). Collectively, these results confirmed that CAR-Ts targeting CD19 have activity in hematologic malignancies and paved the way for studies in a variety of other CD19-expressing B-cell malignancies, mostly NHL histologies.
These studies of CLL also established the role of lymphodepleting chemotherapy with fludarabine and cyclophosphamide (Flu/Cy) by suggesting that depletion of circulating T cells prior to CAR-T infusion increases effectiveness. The trial results also showed that a conditioning regimen of Flu/Cy is superior to single-agent cyclophosphamide conditioning.12-14 The role of lymphodepleting chemotherapy with Flu/Cy was also confirmed by initial NHL studies, in which negative trial results were attributed to lack of conditioning therapy.18
The National Cancer Institute (NCI) conducted a study of CAR-T therapy in 15 patients with advanced B-cell malignancies, 9 of whom had diffuse large B-cell lymphoma (DLBCL); of this group, 4 patients had primary mediastinal B-cell lymphoma (PMBCL).13 Of the entire cohort, 12 patients responded (ORR, 80%) and 8 patients (53%) attained a CR. Of the 7 patients with refractory DLBCL, 4 (57%) achieved a CR with durations ranging from 9 to 22 months. To mitigate treatment-related toxicities, de-intensifying the conditioning program was suggested; however, it is yet unknown how this strategy impacts efficacy.
Kochenderfer et al19 treated 9 patients with B-cell NHL (8 with DLBCL) with low-dose Flu/Cy before CAR-T infusion; one patient with DLBCL attained a CR (13%), and 4 patients showed a partial response (PR; ORR, 62%). In another study that included patients with a variety of B-cell malignancies (18 patients with DLBCL, 6 patients with follicular lymphoma [FL], and 4 patients with mantle cell lymphoma [MCL]), participants were conditioned with either cyclophosphamide alone or Flu/Cy. Flu/Cy-treated patients had a significantly better CR than those treated with cyclophosphamide alone (42% vs 8%), and, among patients with DLBCL treated with Flu/Cy, 38% achieved a CR. Moreover, 2 of 3 patients with FL receiving Flu/Cy conditioning attained a CR.20
Schuster et al21 reported on a trial of 38 patients with refractory B-cell NHL (21 patients with DLBCL, 14 patients with FL, and 3 patients with MCL). All enrolled patients had no curative options and anticipated survival times of < 2 years. Lymphodepleting regimens varied based on disease burden, histology, and previous therapies. Importantly, the median number of prior therapies was 4 (range, 1-10), and 32% of enrolled patients had undergone prior transplantation. The ORR at 3 months was 68% (DLBCL, 54%; FL, 100%; MCL, 50%); at a median follow up of 11.7 months, at the time of this report, progression-free survival (PFS) for the entire cohort was 62% (DLBCL, 43%; FL, 100%).
In the first multicenter trial of CAR-T in relapsed/refractory NHL (ZUMA-1), patients with either DLBCL or PMBCL received 3 days of Flu/Cy conditioning before a single CAR-T infusion. In total, 111 patients from 22 institutions were enrolled; an analysis was recently presented on 51 patients with DLBCL by Neelapu et al.22 The average turnaround time from apheresis to receiving the infusion was 17.4 days. The ORR was 76% (CR, 47%; PR, 29%), with 92% of responses occurring within the first month of infusion. At the time of this report, 39% of patients had ongoing responses at 3 months (CR, 33%). While PFS was 92% at 1 month, 56% of patients were progression-free at 3 months. Only 6 patients with PMBCL were treated in a similar fashion.23 At a median follow up of 3.2 months, ORR was 100% for these patients, and all of these responses were CRs. Strategies to understand how these responses can be sustained are ongoing. However, ZUMA-1 demonstrates that studies of CAR-T therapies can be safely conducted in a multicenter setting, a critical component to ensure marketing and commercial success.