CAR T Cell Meeting 2019 | CAR T-cell therapy in multiple myeloma: the latest clinical data

On 14 February 2019, Alfred Garfall from the Perelman School of Medicine, University of Pennsylvania, PA, USA, presented at the 1^st European CAR T Cell Meeting in Paris, France, on the topic of chimeric antigen receptor T-cell (CAR T) therapy in multiple myeloma (MM). Dr Garfall summarized the latest clinical data, the main downfalls of CAR T therapy in the MM setting and mechanisms of improving efficacy for future constructs and trials. Here, the MM Hub present the main topics from his talk.¹

Immunotherapy and MM:

Immunotherapy is a rapidly evolving field in MM with many potential cell surface targets already identified such as B-cell maturation antigen (BCMA) and CS1 (SLAMF7). Some of these have already been utilized to create clinically targeted licensed therapies such as daratumumab, which is a monoclonal antibody (MAb) targeting CD38.

Whilst the use of CAR T therapy in MM is not currently licensed, there are several ongoing clinical trials, with many more in the pipeline, creating a crowded market. Most CAR T therapies in clinical trials in MM target BCMA, but other trials are currently investigating CS1 (SLAMF7), CD38, CD138, NKG2D ligands, Lewis Y antigen and surface immunogloblulin as potential targets. During ASH 2018, 15 clinical trials targeting BCMA in MM were presented, 13 of which were CAR therapies, which is an incredible number for just one target and therapeutic approach.

Current CAR-T trials in MM:

There are three main anti-BCMA CAR T phase I trials in MM; the latest data for each is listed here:

• Bluebird Bio (bb2121): (N = 43)²
  • CAR T construct: Human CD3/41BB
  • Overall response rate (ORR): 77% (30/39)
  • ORR (optimal doses): 96% (21/22)
  • Dosing: 0.5–8 x 10⁸/kg
  • Median prior lines of therapy: 7.5
  • Cytokine release syndrome (CRS), grade ≥3: 5%
  • Neurotoxicity grade ≥3: 2%
  • NCT02658929
• NCI (CAR-BCMA): (N = 26)^3,4
  • CAR T construct: Murine CD3/CD28
  • ORR: 58%
  • ORR (optimal doses): 81% (13/16)
  • Dosing: 0.3–9 x 106/kg
  • Median prior lines of therapy: 7.5
  • CRS grade ≥3: 23%
  • Neurotoxicity grade ≥3: 12%
  • NCT02215967
• Penn/Novartis (P-BCMA-101): (N = 25)⁵
  • CAR T construct: Human CD3/41BB
  • ORR: 48%
  • ORR (optimal doses): 64% (7/11)
  • Dosing: 0.5–5 x 10⁸/kg
  • Median prior lines of therapy: 7
  • CRS grade ≥3: 32%
  • Neurotoxicity grade ≥3: 12%
  • NCT03288493

Whilst these three CARs all target BCMA and have the most mature data available, there are differences in their design including the lymphodepletion regimen, the pretreatment disease burden, and the need for documented BCMA expression.

All the studies listed above enrolled heavily pretreated patients including those with high-risk cytogenetics. The ORRs therefore are very impressive, considering the population treated. There were also reasonable toxicities in terms of CRS and neurologic events, which was expected given that BCMA is only expressed on a subset of B cells and mature plasma cells, minimizing the off-target effect.

Response Durability:

In anti-CD19 CAR T therapies, which are licensed for the treatment of diffuse large B-cell lymphoma (DLBCL) and acute lymphoblastic leukemia (ALL), the response rates flatten over time indicating that late relapses are relatively rare.

However, in CAR T trials in MM to date, this is not the case. In the Bluebird study, a median progression free survival (PFS) of 17.7 months was seen in patients who were minimal-residual disease (MRD) negative. Whilst this is a significant result in such a heavily pretreated population, it is indicative of the fact that even the patients who have responded best to treatment relapse over time.¹ This was also experienced in the NCI trial whereby patients experienced disease progression over time, following CAR T therapy.⁵

Lessons so far:

So far, clinical trials with anti-BCMA CAR T therapy have shown that BCMA is a valid target for myeloma immunotherapy and that the approach of CAR T is applicable in targets aside from CD19 in the lymphoma setting. Anti-BCMA CAR T can elicit significant responses in the heavily pretreated relapsed/refractory (RR) population, but most patients relapse despite a clinically significant PFS. The cytotoxicity profile is reasonable, with no significant effects on normal tissues, and is in-line with the expected effects from the anti-CD19 experience.

The loss of response:

To understand why there is a loss of response, there are two main questions:

1. Are the CAR T cells persisting?

• Data from Penn Medicine (Cohen et al. in revision) shows initial CAR T expansion followed by persistence over time. Most patients have detectable anti-BCMA CAR T cells at the time of progression

2. Are the CAR T cells active?

• By measuring BCMA expression levels on myeloma cells by flow cytometry at time points post-therapy, it has been shown that CAR T cells persist but lose anti-myeloma activity
• The hypothesis therefore is that a BCMA low (but not negative) phenotype contributes to resistance to anti-BCMA CAR T cells

How can we improve the efficacy of CAR T cells in MM?

1. Provide CAR T during an earlier line of therapy

• Analyzing the immunophenotype of the apheresis product obtained from patients with RRMM showed a specific immunophenotype of T cells was found to predict a higher response to anti-BCMA therapy
  • The immunophenotype identified was CD45R0- CD27+ CD8+ which contains a high proportion of naïve T cells
  • A higher CD4/CD8 ratio also predicted a better response
  • No other factors such as age had any predictive value⁶
• Therefore, increasing T-cell health may increase response rates
• One way to do this is to provide CAR T at an earlier line of therapy:
  • When comparing apheresis products in the post-induction versus RR setting, the post-induction product was fitter and more robust for proliferation⁶
  • Therefore, harvesting earlier in therapy can produce more potent product with a higher likelihood of response

2. Co-target other antigens (CD19)

• MM cells exist in a spectrum between memory B cells (CD19-expressing) and plasma cells (BCMA-expressing). There is some evidence that less differentiated subsets (expressing CD19) may be required for disease regrowth following successful initial cytotoxic therapy
• Therefore, co-targeting CD19 with BCMA may prevent late relapses
• Ex vivo experiments have shown anti-BCMA and anti-CD19 CAR T given in combination leads to more effective depletion of MM cells than anti-BCMA CAR T cells alone.⁷

A new study (NCT03549442), utilizing both strategies listed above (treating with CAR T earlier in the pathway and co-targeting CD19), is currently recruiting patients with high-risk MM. Patients will receive standard first-line therapy and, following a stable response, will have their T cells and stem cells (if transplant eligible) harvested. The initial therapy is then consolidated with either anti-BCMA CAR T therapy or anti-BCMA and anti-CD19 CAR T therapy. Results from a similar study were recently published by Shi et al. at ASH 2018 with promising results providing justification for this approach.⁸

3. Increase the potency of the CAR T cells. This can be done in a manner of ways, including;

• Optimizing the CD4/CD8 ratio and altering the manufacturing process to increase the number of the central memory T-cell phenotype cells
• Altering the gene transfer system to transpose specific stem memory T cells which are less differentiated – has demonstrated early efficacy at a dose of 857 x 10⁶ with an ORR in the three patients treated of 100% at a median 56 day follow up. They also have very low CRS rates allowing the T-cell dose to be increased⁵
• Utilize bivalent BCMA binding – to increase binding in low-BCMA expressing subsets of myeloma cells
• Treating cells in culture with PI3 kinase inhibitor during manufacturing, such as in Bluebird’s bb21217 product where safety data was reported at ASH recently⁹

4. Increase BCMA expression on the cell surface

• BCMA cell-surface expression is regulated post-translationally by extracellular cleavage by gamma secretase which can be targeted by drugs
• Therefore, inhibitors of gamma secretase may increase BCMA expression on the cell surface
• This is a topic discussed during the same congress by Stanley Riddell of the Fred Hutchinson Cancer Research Center¹⁰
  • Dr Riddell presented data showing when a gamma secretase inhibitor was administered, BCMA expression increased
  • Provides preliminary evidence that this mechanism of action may be effective

5. Target multiple plasma cell antigens on the myeloma cell surface

• There are multiple other immunotherapy targets in MM including CS1 (SLAMF7) with many of these are in preclinical trials
• Once they demonstrate safety they may be used in combinations

The immunotherapy field in MM is crowded, especially in comparison to the DLBCL and ALL fields, with many bispecific antibodies targeting BCMA, CD38, and GPRC5D, and antibody-drug conjugates targeting BCMA are also providing promising data.

Therefore, it is currently unknown whether any of these methods or strategies targeting BCMA or other MM immunotherapy targets may preclude another and how they will be utilized. A clinical benefit may be provided in using these in combination or in sequence.