18th International Myeloma Workshop
CARTITUDE-2: Ciltacabtagene autoleucel in patients with progressive multiple myeloma
Prof Adam Cohen - University of Pennsylvania, Philadelphia, USA
This is a multi-cohort, multicentre phase II study using a BCMA CAR T-cell product called ciltacabtagene autoleucel, or cilta-cel. This product was previously tested in other trials – the LEGEND-2 study initially in China and then CARTITUDE-1 which established its efficacy in a US population of very heavily pre-treated patients, triple class refractory. So, based on that initial promising data, CARTITUDE-2 was started and this is now looking at using the same BCMA CAR T-cell product but in often less heavily pre-treated patients and/or in different populations of patients with unmet need.
CARTITUDE-2 right now has five different cohorts and what I was discussing at IMW was cohort A which was looking at using this BCMA CAR T-cell product in patients with one to three prior lines of therapy who were lenalidomide refractory. So that’s the basic background and set-up of the study. It is designed to look at 20 patients per cohort and the primary endpoint for each cohort is looking at the MRD negative rate but obviously we’re also looking for overall response rate, complete response rate, progression free and overall survival and safety as well in these various populations.
As mentioned, patients had to have one to three prior lines of therapy to get on the study and had to be refractory to lenalidomide. Once they enrolled they had T-cells collected by leukapheresis and those cells were then manufactured to express the BCMA CAR. Patients could get bridging therapy during manufacturing, and this was at the investigator’s discretion, to try to maintain disease control during manufacturing. They then received fludarabine and cyclophosphamide as lympho-depleting conditioning followed by a single infusion of the cilta-cel CAR T-cell product. Then they were monitored for safety and efficacy thereafter. Patients had the option of being treated as an inpatient or outpatient to get their CAR T-cells based on the institutional experience and preference.
20 patients were enrolled in the study and this, as planned, was a less heavily treated population than CARTITUDE-1. So in this study there is a median of two prior lines of therapy compared to a median of six prior lines for the prior CARTITUDE-1 study. About two-thirds of patients though were triple class exposed, meaning they had seen a previous proteasome inhibitor, IMiD and CD-38 antibody. 40% were triple class refractory and all of them were lenalidomide refractory as required by the protocol. About a third of patients had high risk cytogenetics going into the study.
In terms of looking at the primary efficacy results, the overall response rate was 95%, meaning 19 out of 20 patients responded, one patient had stable disease as best response. 75% of patients had a CR or better and 85% achieved a VGPR or better. The median time to response was rapid at one month and median time to CR was 1.9 months, so very similar to the CARTITUDE-1 study. Only four patients at this time were evaluable for MRD negativity, all four of them were MRD negative at the 10-5 threshold by next gen sequencing but we’re waiting further follow-up testing to assess the true MRD negative rate.
What was impressive so far, again a relatively short follow-up, only a median follow-up of about six months, but no patient had progressed at the time of data cut-off with the longest patient out about one year from CAR T-cells.
So those were the primary efficacy findings. In terms of safety, overall the safety profile seemed fairly similar to what we’ve come to expect with CAR T-cell therapy in myeloma patients. The primary toxicities included cytopenias which were expected based on the lympho-depleting conditioning although the frequency of high grade, grade 3/4, thrombocytopenia and anaemia seemed a little bit less in this cohort compared to CARTITUDE-1, perhaps because they were less heavily pre-treated. Similarly, there seemed to be a lower incidence of prolonged cytopenias for more than two months.
In terms of cytokine release syndrome, this was seen in 85% of patients but all but two of these were grade 1 and grade 2. There was one grade 3 and one grade 4 and there were no treatment related deaths. The median time to onset of CRS was 7 days and that’s one thing that differentiates this CAR product from some other ones where the median time to CRS is in the order of a day or two. The ICANS or typical correlated neurotoxicity was seen in three patients, or 15%, all grade 1 and grade 2, all resolved after a median duration of two days. Importantly, there were no late neurotoxicity events in this cohort, particularly the Parkinsonian-like movement or neurocognitive treatment related adverse events were not seen in this cohort of 20 patients.
This really just shows that certainly we can give these CAR T-cells in a less heavily pre-treated population of patients. The efficacy seems similar to what’s been seen with their use in later lines of therapy. Perhaps we’re seeing a little bit less of the cytopenias which was one of the hypotheses that when patients were not as quite as beat up from prior therapies they may recover a little bit faster. We’re still obviously waiting for the longer follow-up to see if the duration of responses is going to be longer when we treat less heavily pre-treated patients compared to waiting until they’ve had a median of six or seven prior lines of therapy which has been done for most of the prior CAR T-cell studies.
So we need a little bit longer follow-up but this really shows support for this idea of moving this up as an earlier line of therapy. In fact, these data have been used to support the opening now of a randomised phase III trial for this same population, one to three prior lines, lenalidomide refractory, comparing ciltacabtagene autoleucel, or cilta-cel, versus doctor’s choice of one of two standard of care therapies – either daratumumab pomalidomide dexamethasone or bortezomib pomalidomide dexamethasone. So that phase III trial will really serve perhaps to confirm these data that this is going to be a useful therapy for less heavily treated patients.
I already mentioned the phase III trial that is going forward as a follow-up to this cohort. This cohort is continuing to accrue and there may be another 20 patients in the phase II CARTITUDE-2 programme for cohort A, just to get a little bit more experience. Then obviously we’re continuing to accrue to the other cohorts in CARTITUDE-2 which are exploring some interesting populations of patients, including patients who are progressing within one year of auto-transplant, their initial therapy, that’s a poor prognosis group. There’s a cohort for patients who have had a prior BCMA-directed therapy, so we’ll get a sense of how useful this therapy is in that setting. Then there are two cohorts exploring the use of this really as consolidation for initial therapy, either for patients who have suboptimal response after an autologous transplant or even instead of an autologous transplant but doing this right after induction for high risk patients. So those cohorts are currently accruing and hopefully we’ll have some data over the next year or so exploring the use of cilta-cel in those settings as well.