Test Driving CARS: Optimizing Outcomes

Panelists

• Rebecca Gardner, MD
• Sattva S. Nelapu, MD

Agenda

• Sattva S. Nelapu, MD

CAR T-Cell Development: From Discovery to FDA Approval Over ~25 Years^[1-10]

• The development of chimeric antigen receptor (CAR) T-cell therapy from the initial discovery to US Food and Drug Administration (FDA) approval has taken almost 25 years
  • This concept of CAR T cells was first proposed in 1989
  • The first clinical trial was done in ovarian cancer against a folate receptor in 2006 but was not very effective
  • Subsequent preclinical studies have found that CD19 CAR T-cell therapy appeared to be quite effective in B-cell malignancies
  • The first lymphoma patient was treated at the National Cancer Institute (NCI) in 2009 with CD19 CAR T-cell therapy
  • In parallel, the University of Pennsylvania had been treating patients with leukemia and chronic lymphocytic leukemia (CLL) as well as acute lymphoblastic leukemia (ALL)
  • The success observed in these initial single institution trials led to multicenter trials in ALL and lymphoma, which led to FDA approval of this on August 30, 2017 for ALL and axicabtagene ciloleucel on October 18 for aggressive B-cell non-Hodgkin lymphoma

CAR-Modified T Cells^[11]

• Abnormal T cell recognizes its antigen via its T-cell receptor, and that antigen is presented in the context of a major histocompatibility complex (MHC) peptide complex on the target cell
• CAR T cells: regenetically engineered T cell with an artificial receptor introduced into the T cell
• The CAR has 2 components
  • Extracellular domain
  • Signaling or a cytoplasmic domain

Rationale for CD19 as a CAR T-Cell Therapy Target

• CD19 is expressed across the B cell development, from the precursor B-cell stage to the mature B-cell stage
• ALL and various types of lymphomas, including follicular lymphoma, diffuse large cell lymphoma, as well as other B-cell malignancies, can express CD19

CD19 CAR T Products in Pivotal Trials in ALL and/or NHL

• Of the three CD19 CAR T-cell products undergoing pivotal phase 3 trials, all use the same CD19 antibody clone called FMC63 and use CD3 zeta chain in order to provide signal 1 activation of the T cell
• Signal 2 is different between these products, the NCI product uses CD28 costimulator domain, whereas the University of Pennsylvania (U Penn) and the Fred Hutchinson Cancer Research Center (FHCRC) products use 4-1BB costimulatory domain
• They also use different gene transfer technology
  • The NCI product uses a retroviral technology to introduce the CAR molecule
  • U Penn and the FHCRC/Seattle Children's Hospital (SCH) use lentiviral technology

ELIANA: First Multicenter Trial of CTL019 in Relapsed/Refractory Pediatric and Young Adult ALL^[13]

• The ELIANA trial is the first multicenter trial to evaluate tisagenlecleucel (CTL019) in relapsed or refractory pediatric and young adult ALL
• Eligible patients had relapsed or refractory B-cell ALL after 2 or more prior lines of therapy or primary refractory disease

ELIANA: Patient Disposition^[13,14]

• 68 patients have received the product at this time and follow-up is ongoing in 49 patients

ELIANA: Patient Characteristics^[13,15]

• These were heavily pretreated patient populations with a median number of 3 prior therapies
• A median blast count of 73% suggests that these patients had high tumor burden at the time of enrollment

ELIANA: Efficacy (n = 63) ^[13,16]

• All patients who responded had minimal residual disease (MRD)-negative disease

ELIANA: ORR Across Subgroups^[13,17]

ELIANA: RFS and OS^[13,18]

• Responses were observed across all subgroups

ELIANA: Safety^[13]

• These data led to the FDA approval of tisgenlecleucel on August 30, 2017 for pediatric and young adult ALL
• The most common adverse event is cytokine release syndrome (CRS)
• The second most common toxicity is neurologic toxicity
• All patients who responded did have B-cell aplasia, since CD19 CAR T-cell therapy does not distinguish between tumor B cells and normal B cells

ELIANA: CRS Onset, Duration, and Management^[13]

• To manage the toxicities, one of the most common therapies used is anti-interleukin-6 therapy with the anti-interleukin-6 receptor antibody, tocilizumab

Unmet Need in Refractory Aggressive B-Cell NHL: SCHOLAR Study -- Retrospective NHL Research

• CD19 CAR T-cell therapy was also evaluated in patients with refractory aggressive B-cell non-Hodgkin lymphoma
• The unmet need in this patient population was recently reported in the SCHOLAR study
  • Data from 4 different datasets (2 randomized studies)

ZUMA1: First Multicenter Trial of CD19 CAR T-Cell Therapy in Refractory Aggressive B-Cell NHL^[19]

• ZUMA1 trial is the first multicenter trial of CD19 CAR T-cell therapy with axicabtagene ciloleucel in patients with refractory aggressive B cell non-Hodgkin lymphoma
• Trial schema is very similar to ELIANA trial
• Patients are hospitalized from the time of infusion in order to monitor for toxicity and for at least 7 days after
• Patients who had either diffuse large B-cell lymphoma, primary mediastinal B-cell lymphoma, or transfollicular lymphoma that was chemorefractory

ZUMA1: Patient Disposition^[19]

• The product could be manufactured for 110 or 111 patients for a 99% manufacturing success rate
• The average turnaround time from apheresis to the delivery of the product to the clinical site was 17 days

ZUMA1: Patient Characteristics^[19]

• These data had a cutoff of January 27, 2017, with a median follow-up of about 9 months

ZUMA1: Efficacy^[19]

• One quarter of the patients were aged 65 years or older
• These were heavily pretreated patient populations

ZUMA1: ORR Consistent Across Subgroups^[19]

• Responses could be observed across all subgroups that were analyzed, including the refractory subgroups

ZUMA1: Duration of Response^[19]

• 70% of patients in a complete remission (CR) appeared to have ongoing remissions at 9 months

ZUMA1: Overall Survival^[19]

• The median overall survival has not been reached
• These data led to FDA approval of axicabtagene ciloleucel on October 18, 2017 for patients with aggressive B-cell non-Hodgkin lymphoma after at least 2 prior lines of therapy

Multicenter CAR T NHL Trials: Study Design^[19-21]

• There are additional multicenter trials ongoing in lymphoma
• The JULIET trial is evaluating tisagenlecleucel in patients with aggressive B-cell non-Hodgkin lymphoma and the TRANSCEND trial is evaluating JCAR017 in a similar patient population
• There are some differences in the trial designs
  • ZUMA1 enrolled patients with refractory disease, whereas JULIET and TRANSCEND included patients with relapsed or refractory disease

Multicenter CAR T NHL Trials: Efficacy and Safety^[19-21]

• The best overall response rate across the 3 trials was in the range of 60% to 80%
• The CR rates ranged from approximately 45% to 55%
• Grade 3 or higher CRS was observed in 13 patients in ZUMA1, 26 patients in JULIET, and 2 patients in TRANSCEND; grade 3 or higher neurologic toxicity was observed in 28%, 13%, and 16% of patients, respectively

ZUMA1: Tocilizumab/Steroid Use: Impact on Clinical Efficacy^[19]

• Interventions used to manage these toxicities included the anti-interleukin-6 antibody, tocilizumab, and corticosteroids
• When outcomes in patients receiving these interventions were compared with those who did not receive them, there was no significant differences in the overall response rate, CR rate, or the durability of responses observed
• The interventions appear to, at least in the case of steroids, temporarily lower the CAR T-cell numbers in order to manage the adverse toxicity

ASH 2017: CAR T Abstracts of Interest

• A number of important abstracts will be presented at the American Society of Hematology (ASH) meeting

CD19 CAR T: Key Takeaways

CD19 CAR T in Pediatric ALL: The Beginning of a Paradigm Shift

• CD19 CAR T cell therapy is just the beginning of a paradigm shift of how we will be treating B-cell malignancies
• We may be able to cure 95% of pediatric ALL

CD19 CAR T in NHL: The Beginning of a Paradigm Shift

• For newly diagnosed large cell lymphoma, approximately 80% of those patients may be curable using these 3 modalities
• This therapy is now being evaluated in other B-cell lymphomas, including indolent B-cell lymphoma as well as mantle cell lymphoma
• Early next year, we will be seeing randomized trials where the CD19 CAR T-cell therapy is going to be compared directly with efficacy against autotransplantation in the second-line setting as well as evaluating the efficacy of high-risk aggressive B-cell non-Hodgkin lymphoma in the frontline setting

Pearls to Improve Patient Outcomes

Case 1: Mike

• Mike was treated up front with rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone (R-CHOP) chemotherapy for 6 cycles and intrathecal methotrexate
• He obtained a CR, which lasted for 13 months
• He was found to have progressive disease in the mediastinum, then received 3 cylcles of rituximab, ifosfamide, carboplatin, etoposide (R-ICE) chemotherapy, again obtained a CR
• He went on to receive consolidation with carmustine, etoposide, cytarabine, melphalan, and rituximab followed by autologous hematopoietic stem cell transplantation, and remained in remission for 2 years
• He had a growing lymph node in his neck and was biopsy-proven to have progressive diffuse large B-cell lymphoma; at the time he had B symptoms as well
• He also progressed on 3 cycles of rituximab, gemcitabine, oxaliplatin (R-GEMOX)

Panel Discussion

• Is this patient eligible for CAR T-cell therapy?
• Dr Neelapu: Yes. Until recently, patients who relapsed or progressed after autotransplant or were refractory to salvage therapy did not have an established standard of care
  • Axicabtagene ciloleucel (Axi-cel) is FDA approved in this setting
• Dr Locke: The earlier we refer these patients, the better off they will be
• Rebecca Gardner, MD: We think that the earlier, the better the T-cell collection

Case 1: Mike (cont)

• Mike's T cells were collected via leukapheresis for Axi-cel CAR T-cell therapy
• The evening of his infusion he developed a fever of 38.2°C, and broad-spectrum antibiotics were administered; again, back to the panel
• What is the importance of giving conditioning chemotherapy in a case like this?
  • Dr Gardner: Yes. The use of the chemotherapy is to create a lymphodepleted environment in which your homeostatic cytokines are increased. Then, when you infuse your CAR T-cell product, it is a favorable environment for those CAR T cells to be able to expand and proliferate in the patient.
  • Dr Neelapu: In addition to lymphodepletion, this conditioning chemotherapy likely acts by other mechanisms, including depleting various immunosuppressed cells like regulatory T cells and myeloid suppressor cells. In addition, conditioning chemotherapy can disrupt the mucosal barrier and that can lead to better activation of antigen-presenting cells and better activation of T cells in the process.

Case 1: Mike (cont)

• Mike developed a fever the day of infusion
• The fever continued to escalate
• He also had sinus tachycardia, but otherwise his vital signs were stable over a number of days, up to day 4
• 5 days after his CAR T-cell infusion he developed a grade 3 CRS characterized by hypertension, atrial flutter, and worsening tachycardia
• Tocilizumab was administered

Panel Discussion

• What are the signs and symptoms of CRS?
  • Dr Gardner: CRS is a multifaceted syndrome that we frequently see after CAR T-cell therapy. The initial presenting symptoms are tachycardia and fever, just as we see in our patient case here. The fever can become quite high into the range of upper 39^o to 40^oC, and higher, and can persist for many days.
    • The most predominant severe symptoms that we see is hypotension, but others include respiratory and renal dysfunction and progression to other organ dysfunction
  • Dr Neelapu: CRS is a systemic inflammatory response
    • There are multiple different cytokines produced. The central mediators of CRS symptoms, as well as multiorgan dysfunction, appears to be interleukin-6, and that is the reason the anti-interleukin-6 receptor antibody, tocilizumab, is used for the management of CRS.
    • There are also other cytokines which may be involved. If the patient doesn't respond to anti-interleukin-6 therapy, we may have to use corticosteroids to globally suppress multiple different cytokines.
  • Dr Locke: I always tell patients who are about to undergo CAR T-cell therapy that you can expect to feel pretty crummy. You are going to have a high fever for many days.
• What is the general timing of the onset of CRS?
  • Dr Neelapu: The onset does vary from product to product and depending on the dose of CAR T cells infused. In the case of the ZUMA1 trial, with Axi-cel, the median onset time for CRS was 2 days, with a range of 0 to 8 days. The median duration of CRS was 7 to 8 days.
  • In the ELIANA trial, the median onset time for CRS was 3 days, but sometimes the onset can be later.
  • You have to educate the patient as well as the family. Any fever that occurs within the first 1 to 3 weeks after the CAR T-cell infusion needs to be evaluated for possible CRS.
• Is there anything different in the pediatric population?
  • Dr Gardner: In general, for the ALL patients, we believe that the onset of CRS occurs a little sooner than for patients with NHL
    • The pediatric patients tend to have less organ dysfunction with their CRS, probably because they generally have fewer comorbidities at the start
• How do you grade CRS?
  • Dr Gardner: The grading of CRS is a little bit complicated right now. Typically, we rely on Common Terminology Criteria for Adverse Events (CTCAE) for grading of adverse events (AEs). If you read the definition of CRS in the CTCAE, it is infusional-related CRS, which is not typical for what we see.
    • Some institutions have developed their own grading systems. The Lee criteria was published a few years ago.
    • It is very important when you are looking at studies and comparing across trials that you are cognizant of these differences in CRS grading, because it affects how you view severe toxicity and what interventions should be used
• When should we intervene and treat CRS?
  • Dr Neelapu: There are 4 parameters I evaluate
    • 3 are vital signs: temperature, blood pressure, and O₂ saturation; the fourth parameter is evidence of organ toxicity
• Does tocilizumab or other interventions have an impact on the efficacy?
  • Dr Neelapu: Tocilizumab is currently used for grade 2 or higher CRS, and occasionally neurotoxicity as well
    • For corticosteroids, we generally use it for grade 3 or higher toxicity
    • Initially, there were many questions whether tocilizumab would affect efficacy, but data with a variety of products in a variety of disease settings suggest that it does not affect efficacy
  • Dr Gardner: The clinical symptoms dictate your intervention. We should be trying to look at intervening earlier to prevent more severe toxicities.

Case 1: Mike (cont)

• Mike had a grade 3 CRS
• He was transferred to the intensive care unit (ICU)
• He was treated with 3 doses of tocilizumab (8 mg/kg)
• His CRS was downgraded to grade 1
• On day 7, he started to have mental status changes, word-finding difficulties, aphasia, confusion, not being able to feed himself, or perform his activities of daily living (ADLs)

Panel Discussion

• What are the general signs and symptoms of neurotoxicity?
  • Dr Gardner: Neurotoxicity typically occurs after the height of your CRS, and the initial symptoms can be pretty subtle
    • Those symptoms can progress to more severe cases where you have a grade 3 or 4 encephalopathy, including complete aphasia, and at times your patients present with seizures
  • Dr Neelapu: The onset of neurotoxicity can be biphasic. It can occur during the CRS phase, but it can also occur after the CRS phase has subsided.
    • Both product labels include a warning that the patient should stay close to the hospital, within 2 hours driving distance, at least for the first 4 weeks, and should not operate any heavy machinery or drive cars for the first 8 weeks
• How do we grade this neurotoxicity and when should we intervene?
  • Dr Gardner: Currently, grading of neurotoxicity is based on the individual symptoms of the neurotoxicity
  • Dr Neelapu: MD Anderson has developed a coding system sort of like a Mini-Mental State Examination score where patients are asked to perform 10 different tasks or answer questions. For each correct answer, you give it a point, and depending on how much the score, you can assign a neurotoxicity score grade.
    • Two of the earliest signs that we observed for neurotoxicity were impaired handwriting, and the ability to count numbers backwards
• Dr Locke: Are there any other toxicities that we need to think about with these therapies?
  • Dr Gardner: One of the other toxicities we frequently see in the ALL patients is development of disseminated intravascular coagulation
    • This usually occurs after CRS
    • The other long-term toxicity we see is depletion of normal B cells

Case 1 Follow-Up: Mike

• Mike was treated withdexamethasone, 10 mg intravenous (IV) every 6 hours, resulting in stabilization of his mental status
• His mental status improved over that time
• Mike remains in CR
• He has B-cell aplasia, had several prolonged viral upper respiratory infections, and was started on monthly intravenous immunoglobulin (IVIG)

Panel Discussion

• Do you guys put all of your patients on IVIG?
  • Dr Neelapu: I monitor these patients for the immunoglobulin G levels, but generally do not start IVIG unless they develop recurrent infections
  • Dr Gardner: For the pediatric ALL population, we are repleting everybody who has an immunoglobulin G level below 400 and we will initially start repletion with IVIG. If they have ongoing persistence, many times we switch them to subcutaneous immunoglobulin.
  • Dr Locke: For adults, we generally follow for recurrent infections and initiate IVIG, if that occurs in the setting of hypogammaglobulinemia
• Dr Locke: In lymphoma, we do not always see CR by 1 month
• Dr Neelapu: Even if a patient has stable disease or partial response, it is appropriate to watch those patients and see if that response deepens

Case 2: Sheila

Panel Discussion

• The majority of the audience chose number 3, stop and evaluate the patient for infection and inflammatory state
• What would you do in this situation?
• Dr Gardner: CRS creates a highly inflammatory state. In order for a patient to be able to optimally handle that state, it is important that they are not already inflamed. We have evidence that if you are already inflamed, the time you get your CAR T cells, your toxicity will be worse and potentially fatal.
• Dr Neelapu: It is now a black box warning for both approved products. If the patients have active infections, you should not proceed with CAR T-cell therapy. Also, if the patient has an active inflammatory disorder, such as an autoimmune disorder, those patients also should not get a CAR T-cell infusion until that has been controlled and evaluated.
• Dr Locke: In the phase 1 portion of the ZUMA1 trial, 1 of the 7 patients died shortly after infusion of CAR T cells. The patient had a fever prior to initiation of cyclophosphamide/fludarabine chemotherapy
  • The patient had elevated inflammatory cytokines, and rapidly progressive disease

Case 2 Follow-Up: Sheila

Looking to the Future

• Rebecca Gardner, MD

Pitfalls of the Current State^[13,19-21]

The Future of CD19 CAR T Cells^[9,10,22-25]

• The rationale for a movement toward humanized CAR T cells is that some patients lose engraftment of their CAR T cells and have developed a rejection response against the murine-based antibody of the CAR construct
• Multiple studies have shown that persistence of the CAR T cell matters for durability of remission, particularly in patients with ALL

Overcoming Barriers With Additional Engineering^[26]

• There are additional ways to engineer the T cells to make them inherently better
• The checkpoint blockade is an area of interest and has stand-alone efficacy in lymphoma

Overcoming Barriers With Additional Engineering (cont)^[27]

• There is a movement to create an off-the-shelf product

Overcoming Barriers With Additional Engineering (cont)^[28]

• CD19 CAR T-cells can develop antigen escape
• It will become more important that we have dual targeting of antigens
• Pooled CAR T-cell product would combine CAR T cells against 2 different antigens infused simultaneously
• Another strategy is a multi-CAR T cell in which one T cell had multiple expressions of different CAR constructs on the surface
• Tandem CAR T-cells would have 2 antibody domains linked together with 1 intracellular domain

Overcoming Barriers With Combination Therapy^[29-32]

• Another strategy is to combine the CAR T-cell product with another type of therapy

Future Directions

• There are a number of other targets being examined

Ongoing Studies in Other Disease States^[33,34]

• In multiple myeloma, B-cell maturation anti (BCMA), which is restricted to certain types of B cells, including healthy plasma cells, is also expressed by the malignant plasma cells that most patients with multiple myeloma have
• Several studies with early results presented for using BCMA CARs had an overall response rate (ORR) around 90% to 100%
• There was similar toxicity with CRS, but potentially at lower rates
• This is a positron emission tomography scan of a patient who had significant disease burden prior to treatment and a very rapid response

AE Grading in Development

• CRS grading is not standardized
• Some studies will call grade 3 and higher severe CRS, and some studies will call grade 4 severe CRS
• This Table is unpublished institutional data from the phase 1 CD19 CAR T-cell trial in ALL in pediatrics
• CRS was graded based on 3 different criteria
  • The grading results were variable

Neurotoxicity: Grading^[35]

• There appears to be a higher frequency of treatment-related neurotoxicity in ALL compared with other diseases. There has been speculation about whether it is related to the specific CAR construct
• Neurotoxicity occurs with or without fludarabine-containing regimens

Neurotoxicity: Grading (cont)^[35]

• Dr Neelapu's group has proposed a new grading system for CAR T-related neurotoxicity
• The system is called CAR-related encephalopathy syndrome

Overcoming Toxicities^[36-38]

• Early interventions included giving patients tocilizumab for persistent symptoms of mild CRS, and then dexamethasone if tocilizumab was not efficacious
  • We were able to decrease the rate of severe CRS by 50%
  • It did not affect the CAR T-cell engraftment or the efficacy of the T cells
  • However, it did not decrease the rate of neurotoxicity or the rate of severe neurotoxicity
  • You can also engineer some safety switches into your CAR T cells

Key Takeaway Messages

ASH 2017: CAR T Abstracts of Interest

Thank you for participating in this activity.

This content has been condensed for improved clarity.