January 2018 Edition Vol.11, Issue 1

CAR-Ts on the Move: Hematologic Immunotherapy at ASH 2017

By Neil Canavan

It’s rush hour in the world of chimeric antigen receptor (CAR) T cell therapy.  Every CAR-T presentation at the annual meeting of the American Society of Hematology (ASH) was packed – and there was a plethora of presentations. Highlighted herein are the some of the most prominent. (In extreme brief, a CAR-T cell is a genetically engineered, tumor-targeted, T cell.)

Kymriah — First CAR-T Cell Approved

Kymriah [tisagenlecleucel; Novartis], indicated for the treatment of children with acute lymphoblastic leukemia, targets CD-19-expressing B cells. Looking to extend that mechanism of action to other indications, Stephen Schuster, MD, Director of Lymphoma, Penn Medicine, reported a primary analysis of the phase II, JULIET trial, using Kymriah in the setting of relapsed/refractory (R/R) diffuse large B-cell lymphoma (DLBCL).

“Patients with DLBCL that have failed available therapies have a very poor prognosis,” said Dr. Schuster. “Response rates with salvage therapy is about 8% for a complete response (CR), and about 18% for partial response, but these tend to be short lived, and the median survival for these patients is four months.”

Results for JULIET, a multicenter, global trial of 81 DLBCL patients were impressive; the overall response rate (ORR) with tisagenlecleucel at one-month was 53%, with a complete response (CR) of 40%. At 3 months, the CR was 32%; at 6 months (n=46), it was 30%. Most importantly, the probability of being relapse-free at 6 months was 73.5% (Figure 1).

 

Figure 1. JULIET: Primary Endpoint Was Met, ORR 53%

 

“I anticipate that, like the (phase I) trial from Penn, we will see similar durable remissions that will last for years,” said Dr. Schuster, adding, “keep in mind, this is in response to a single treatment.”

As for adverse events, suffice it to say that CAR-T cells have two unique side effects, that of the so-called cytokine release syndrome (CRS), which in some early CAR-T trials has been fatal, as well as neurologic toxicities, which are common, but rarely life threatening.

The adverse events reported for JULIET were consistent with the above generalization, and there were no drug-related deaths reported.

JCAR017 in Large B Cell Lymphomas

Chasing a very similar indication as tisagenlecleucel, a flurry of results was presented for the CD-19-targeting product: JCAR017 [Juno Therapeutics], in R/R, large B cell lymphomas, as reported by Jeremy Abramson, MD, Clinical Director Lymphoma, Massachusetts General Hospital.

JCAR017 was tested at three dose levels, over three mixed-type, large B cell lymphoma patient cohorts, which was expanded in two mixed lymphoma cohorts, and one “pivotal” DLBCL cohort.

Out of all these data, the significant take home points for these heavily pretreated patients – half of whom had never achieved a CR – are these:

  • For the entire dataset, at one month the ORR was 74%, and the CR 52% (N=91)
  • At 3 months: the ORR was 53%, the CR, 44% (n=72)
  • At 6 months: the ORR was 35%, and the CR was 31% (n=54)

Across all JCAR017 dose levels, 80% of patients in CR at 3 months remained in response at 6 months, “and 92% of patients in CR at 6 months stay that way,” said Dr. Abramson.

For all 91 patients evaluated, the rate of CRS was “low” at 35%, with only one reported serious case, which resolved. The reported 12% incidence of severe neurotoxicity was “almost entirely reversible.” Both events occurred within the first 10 days after treatment.

Based on these data, Juno is carrying forward the fixed dose of 100 million JCAR017 cells for the ongoing enrollment of the pivotal DLBCL cohort.

 

Anti-BCMA CAR-T Construct: bb2121

The two CAR-T products mentioned thus far, JCAR017 and tisagenlecleucel, as well as the second commercially available CAR-T drug – not discussed here – Yescarta [axicabtagene ciloleucel; previously KITE, now Gilead] exploit the proof of principle target for CAR-T technology, the CD-19 receptor. Thus, there was great excitement over the reported efficacy of targeting a second B cell antigen, the so-called, B-cell maturation antigen (BCMA).

Reporting updated trial results for bb2121, an anti-BCMA CAR for the treatment of multiple myeloma, was James Kochenderfer, MD, of the National Cancer Institute. “Currently, multiple myeloma is a disease that is essentially incurable,” said Dr. Kochenderfer, “So new treatments are desperately needed.”

The data reported were for a multicenter, dose escalation study with doses ranging from 50 million, to 800 million anti-BCMA, CAR-T cells in a population of R/R multiple myeloma patients. “One very critical thing about this study is how heavily pretreated these patients were,” said Dr. Kochenderfer. “The median line of treatment was 7, and all patients had had a prior stem cell transplant.”

The results for treatment with bb2121 were impressive (N=21). The ORR for the entire cohort was 89% at one month, and a full 100% for those patients treated with doses of 150 million or more cells. The rate of CR was 56%. Further, no patients treated at ≥150 million cells experienced disease progression beyond the eight-week time point – with some patients out as far as 54 weeks. There was also evidence suggesting a deepening response over time.

As for adverse events, “Generally, this was a well-tolerated CAR-T cell product,” said Dr. Kochenderfer, “Especially in comparison to other protocol I’ve participated in.” There were only two cases of serious CRS reported, and only one case of serious neurotoxicity. All three patients recovered.

 

Pricing a new CAR (T cell)

Hand in glove with the commercial launch of CAR-T cells was a cost-benefit analysis to defend their enormous expense. The first of these, detailed at ASH, was the rationale for the proposed price point for the Novartis drug, Kymriah, as presented by Yanni Hao, PhD, of Novartis.

“The question is – how do you evaluate the worth of an intervention?” asked Hao. “Do you do it from the perspective of the patient? The payer – the institution – society at large?”

With input from economic modeling experts from Tufts Medical Center, and Johns Hopkins, and considering algorithms used by American Society of Clinical Oncology, and the National Comprehensive Cancer Network, a “willingness to pay” (WTP) by a patient, payer, or society was set at $150,000 per quality of life adjusted year (QALY) gained. This method is similar, Hao pointed out, to the way the British health agency, NICE, judges the economic merits of new medications.

This cost-efficacy calculation was based on the proportion of patients alive, as defined as the OS curve; the total proportion of patients represented by the event-free survival curve; and the number of patients in the progressive/relapsed disease state, at a given time point, as compared to historical controls.

The analysis considered a time-from-treatment horizon of 20 years (a modest span for a drug indicated for a pediatric patient population).

“Over a 20-year time horizon,” reported Hao, “Kymriah, as compared with other treatments was associated with a 2.3 to 4.3 increase in QALYs,” which would justify a price tag in the range of $600k to $779k.

In that light, Novartis set the list price for Kymriah at $475,000 – well under the calculated range. Further, Hao pointed out, if the patient has no response to treatment at one month there is no charge at all.

Hao admitted, however, that the cost-benefit analysis has two obvious limitations – there are no long-term data for Kymriah beyond a few years, and the calculation did not consider the not-insignificant number of cases where patients with suboptimal responses required retreatment with Kymriah.

 

Business Upgrade?

There are two significant obstacles to the wider commercialization of CAR-T cells. The first being that all CAR-T companies of notable size (with the single exception of Cellectis) are using autologous product –cells derived from the individual patient – making the product only suitable for that one patient. From the business perspective, a drug cannot be mass-produced because it is so personalized, and is a fixed cost of goods problem.

Further, the manufacturing of a CAR-T cell product takes several weeks at a centralized facility – for critically ill patients, this is also a problem.

Enter the potential of generating “universal” CAR-T cells by using induced, pluripotent stem cells (iPSCs) derived from donor T cells (TiPSCs).

“The idea is to go from using autologous, patient-specific treatment to the more off-the-shelf allogeneic direction,” said iPSC study presenter, Sjoukje van der Stegen, Research Fellow, of Memorial Sloan Kettering, New York. “If we have our iPSC bank we can just make these cells in unlimited quantities, and if a patient needs it, it can be ordered, and the patient can be treated within a matter of days, rather than weeks.”

The science of generating iPSCs has been known for some time, what is new is the genetic engineering tool, CRISPR, that enables refinement of iPSC products.

“With T cell receptor knockouts (TCR; the current investigative approach) allogeneic GVHD is a concern,” said van der Stegen. (GvHD – graft vs. host disease – is a potentially life threatening side effect of donor-derived, stem cell transplants). There will always be the chance that some level of TCR remain. “But with iPSCs you can guarantee the resulting T cells are 100% TCR-negative if you start with an iPSC clone that was TCR negative.”

The science in brief: Remove T cells from a healthy donor and reprogram them – induce them – via cytokines, to revert to a more primitive, stem cell-like, regenerative state (TiPSC). Then drop the DNA sequence for a CAR in at the TRC-encoding chromosome region (thereby disabling the potential for GVHD due to endogenous TCR expression) and then differentiate according the T cell type(s) required for the “off-the-shelf” therapeutic product.

As presented at ASH, these tricks have already been performed in vitro, and preclinical work is about to commence. Again, from a business perspective these reproducible results are already encouraging – after all, first you want the green pill to work for everyone, and second, you want it to be the exact same green pill every time you make it.

“It’s still biology, so we don’t know yet to what extent it will be the exact same shade of green every time, but there will be far less variability than there is now between every patient,” said van der Stegen, “And I hope it would also allow for easier centralized production.”

Van der Stegen’s optimism for the TiPSC approach is well founded – she is currently a research fellow in the laboratory of Michel Sadelain, one of the scientific founders of Juno Therapeutics.

 

About the Contributor

Neil Canavan is the author of the recently published book, A Cure Within: Scientists Unleashing the Immune System to Kill Cancer.

Post a Comment