Ajai Chari, MD, discusses selecting between CAR T-cell therapies and bispecific antibodies in multiple myeloma, expands on the factors that can help inform these decisions, and highlights the need for additional data to help inform sequencing and potential combinations for these therapies.
Although the BCMA-directed CAR T-cell therapies idecabtagene vicleucel (ide-cel; Abecma) and ciltacabtagene autoleucel (cilta-cel; Carvykti), the BCMA-targeted bispecific antibodies and teclistamab-cqyv (Tecvayli) and elranatamab-bcmm (Elrexfio), and the GPRC5D-targeted bispecific antibody talquetamab-tgvs (Talvey) have been approved by the FDA for the treatment of select patients with relapsed/refractory multiple myeloma, the investigation of additional agents and novel targets such as FcRH5 could help expand the arsenal of T-cell redirection therapies, according to Ajai Chari, MD.
“For a relatively rare cancer, we already had a fair number of [treatment] choices in myeloma. However, they were limited by their duration of efficacy,” Chari explained in an interview with OncLive® during the 2023 SOHO Annual Meeting.1 “Now, we have this huge explosion in interest in T-cell redirection, [with] multiple CAR T-cell therapies and bispecific antibodies. The question is: How do we use all these? How do you pick among [the available] CAR T-cell therapies and bispecific antibodies, and where do these belong among conventional therapies?”
In the interview, Chari discussed his presentation on selecting between CAR T-cell therapies and bispecific antibodies in multiple myeloma, expanded on the factors that can help inform these decisions, and highlighted the need for additional data to help inform sequencing and potential combinations for these therapies. Chari is the director of clinical research in the Multiple Myeloma Program at the University of California, San Francisco (UCSF) Health, UCSF Helen Diller Family Comprehensive Cancer Center.
Chari: Currently, [the CAR T-cell therapies and bispecific specifics] are [approved for patients who received] 4 or more lines of therapy and are triple-class exposed, meaning a proteasome inhibitor, an immunomodulatory agent, and an anti-CD38 [monoclonal antibody]. That applies to all the accelerated approvals [for CAR T-cell therapies and bispecific antibodies] we've had in the last [few] years. Assuming a patient meets those criteria, the next biggest determinant is the rate of their disease progression. The biggest differentiation between bispecific and CAR T is the vein-to-vein time for CAR T [due to] apheresis, manufacturing, and administration, vs bispecifics being off the shelf.
It's important to remember that in almost all of the CAR T studies that have been done up to now, particularly those initial phase 1/2 trials in heavily pretreated populations, there was a long waiting list of patients with limited [CAR T-cell manufacturing] slots. Therefore, when an institution got a slot, you could cherry pick your best patient. That's a selection bias we've never had in myeloma, or probably in any disease. Historically, you test in relapsed/refractory disease. You get consent and you [start treatment] very soon thereafter.
That selection is important in the real-world, as well. If you have a patient who is rapidly progressing, will they be stable for apheresis? First, you must get insurance approval, then get your apheresis slot, which may be several weeks away. After apheresis is done, you still have to keep the patient stable for at least 4 to 6 weeks, and possibly longer. Now, we are also having more and more of these CAR T-cell therapies come in “out of spec.” This means that the product, while probably efficacious, doesn't meet the prespecified cutoff [for CAR T cells] that was discussed with health authorities [when these products were approved]. We don't have long-term follow-up on those patients [who received out-of-spec CAR T-cell therapy].
When you put all that together, if you’re in the clinic and a patient has disease that is [rapidly progressing], I would steer that patient toward a bispecific. Conversely, if a patient has relatively indolent disease, they live quite far away from an academic center where ongoing bispecific treatment may be a hassle, and they can tolerate high-grade cytokine release syndrome [CRS] or neurotoxicity—maybe not the most frail patient, but a fit patient—I would prefer a CAR T-cell therapy for that patient, because they are going to have an amazing treatment-free interval that we can never provide with almost any other therapy.
Currently, we have 2 CAR Ts to select from: ide-cel and cilta-cel. They were both studied in heavily treated patient populations with similar triple-class refractory [disease]. We have been impressed with cilta-cel’s efficacy, with an overall response rate [ORR] of [98%] and a [median] progression-free survival [PFS] of [34.9 months] in the [phase 1/2 CARTITUDE-1 trial (NCT03548207)].2
We always say to have caution with cross-study, single-arm comparisons. With ide-cel, the median PFS was [12.2] months [in the phase 2 KarMMA trial (NCT03361748)].3 [The ORR] and complete response rate were less [than cilta-cel]. Now, we have phase 3 data comparing [cilta-cel and ide-cel] vs a control arm [in the CARTITIUDE-4 (NCT04181827) and KarMMa-3 (NCT03651128) trials, respectively], where the hazard ratio [for PFS] for ide-cel was 0.49 [95% CI, 0.38-0.65; P < .001], and cilta-cel was 0.26 [95% CI, 0.18-0.38; P < .001].4,5 The control arm did better [in CARTITUDE-4 vs KarMMA-3], so in spite of having a better control arm, [cilta-cel demonstrated] a deeper PFS improvement, speaking to the unprecedented efficacy of cilta-cel. I would say to physicians, if you can, get cilta-cel. There still is a role for ide-cel where you may not have a cilta-cel [manufacturing spot] for a patient who is BCMA-naïve who lives far away.
With bispecifics, we have 2 that are now approved [that are] BCMA[-targeted agents]: teclistamab and elranatamab. Both agents have step-up doses and hospitalization requirements after the step-up dosing. In general, the BCMA class is the most crowded for bispecifics. [Regarding] efficacy and safety—again these were small arms—we're not seeing major differences [between the 2 agents]. [Choosing between them] will probably come down to convenience, schedule, and cost. We have never really had the opportunity to consider [these factors] much in myeloma, but it's great for patients to have more products available.
The final bispecific that was recently approved is talquetamab, targeting GPRC5D. The efficacy is quite good for both [BCMA- and GPRC5D-targeted bispecifics]. Where you see the differentiation [between these agents] is with infection. BCMA[-targeted bispecifics are associated with higher rates of] grade 3 or higher infections with long follow up. These are unusual, opportunistic infections that we've never seen before. What I always encourage people to look at with BCMA-targeted bispecifics is non-progression deaths. If the primary cause of death in a study is not progression, then we really need to understand the safety. Because we don't yet have randomized phase 3 trials for bispecifics, we are in a bit of a data gap.
With talquetamab, you don't see [as many] infectious deaths and deaths due to adverse effects [AEs]. That’s nice, but instead, you have different AEs that we're learning how to manage. They can be broken down into skin, nail, and oral toxicities. Rashes or palmar plantar erythrodysesthesia can be managed by topical or oral steroids. For [managing] nail toxicities, we still haven’t found the magic bullet. However, the most challenging [AE to manage] is oral toxicity, which can manifest with taste loss, difficulty swallowing, and weight loss. You can encourage hydration and salivary substitutes, but the mainstay is dose modification.
Those are the current assets that we have, and it's nice to have choices. Being able to sequence some of these agents [could address] another unmet need.
One of the things that we've never done a good job with in myeloma, because we’re [focused] on getting drugs to the patients, is how to thoughtfully use drugs in combination and sequences. This is even more of a question [in the CAR T-cell therapy and bispecific era]. Some things to keep in mind when we think about these assets: What happens after [treatment CAR T-cell therapy or bispecifics]? Some of the issues to think about are antigen and genomic loss. Is the target is still expressed at a protein level and at a genomic level? We are now seeing papers [suggesting] that if you have monoallelic loss at baseline prior to BCMA- or GPRC5D[-directed therapy], there is potential for getting biallelic loss. If a patient has lost BCMA or GPRC5D, then in that subsequent therapy, you might want to switch targets.
Another big thing that we are understanding is the importance of T-cell exhaustion and T-cell fitness. For example, with cilta-cel, which has an amazing [median] PFS of [34.9 months in CARTITUDE-1], patients [treated in the phase 2 CARTITUDE-2 trial (NCT04133636)] who had a prior BCMA bispecific [had a median] PFS that drops to [5.3] months. That's a big warning shot that if you have a patient that you're thinking about for cilta-cel, don't use a BCMA-targeted bispecific right before.
Whether this also applies to GPRC5D-directed bispecifics is unclear. Currently, I'll put patients through apheresis so that the T cells I've collected are good. Then, I could maybe use talquetamab during the bridging period. However, it's very important that we drill down now and not just lump all pre–CAR T-cell therapy treatment into one continuous segment. It is pre-apheresis and post-apheresis. If you're putting all your eggs in that cilta-cel basket, you have to make sure that the T cells that you're collecting are as good as possible.