June 2019 Edition Vol.11, Issue 6

Cell Therapy Rescues ASCO 2019 & Makes Solid (Tumor) Progress

By Neil Canavan

In music, if the first album is a masterpiece, anything that follows will almost certainly disappoint; this has much to do with why the cancer immunotherapy (IO) presentations at ASCO 2019 were so ho hum. The immuno-oncology (IO) debut generated a lot of excitement on the profound efficacy of anti-CTLA-4 and anti-PD-1 therapies in the melanoma and lung cancer settings. This excitement was followed by CAR-T cell treatments for acute leukemia and lymphoma.

At ASCO 2019, the subject of two tumor infiltrating lymphocytes (TILs) therapy presentations by Iovance, showed efficacy not only in metastatic melanoma, but responses were also seen in advanced cervical cancer.

TIL now

In brief, TIL—living cells that are making an impressive impact in solid tumors—is an autologous cell product, a therapy consisting of T cells derived from a patient’s tumor bed that have been isolated and expanded ex-vivo before being infused back into the patient. There’s no genetic manipulation like with CAR-Ts, nor is there cell selection for targeting a specific antigen prior to administration. TILs attack the cancer according to the intrinsic instructions of the patient’s own immune system. The TIL approach works by ramping up the attack on cancer cells.

Amir Jazaeri, MD, of the M.D. Anderson Cancer Center, presented the results of innovaTIL-04, a phase II investigation of the TIL product, LN-145, in the setting of recurrent, metastatic, or persistent cervical cancer.

“This is the leading cause of cancer-related death in women,” said Jazaeri. “There are 12,000 new cases each year in the U.S., and 4,000 related deaths.” The unmet medical need is clear. Once the disease has advanced, the progression free survival (PFS) is only 8 months after front-line chemotherapy; and second-line therapy with pembrolizumab has an overall response rate (ORR) no higher than 14%. Median overall survival (OS) is 8.4 months.

Early investigations with TILs in this setting showed that (a) TILs were present in HPV-associated cancers (cervical, being one); (b) the presence of TILs in the tumor bed was prognostically favorable; and, (c) TILs could be reliably isolated from cervical cancer patients. A previous pilot study of TILs in this setting demonstrated an ORR of 33%, with two complete responses (N=9).

For the current investigation, cervical cancer patients were treated—after TILs were harvested, isolated, and expanded (a 22-day turnaround)—with a preconditioning, lympho-depleting chemotherapy regimen (cyclophosphamide, fludarabine) followed by administration of LN-145, followed by up to six doses of the T cell-boosting agent, IL-2 (600,000 IU/kg; N=27).

Adverse events for this regimen were “tolerable” with the most common grade 3/4 events being anemia (55%), thrombocytopenia (44%), and febrile neutropenia (29%)—largely due to the chemo preconditioning.

As for efficacy, 3 complete responses (CRs), 9 partial responses (PRs), and 11 patients with stable disease were observed (Figure 1). “At a median study follow up of 7.4 months, the median duration of response has not been reached,” said Jazaeri.

Following this presentation, and the melanoma data detailed below, Iovance stock shot up 50%.

 

Manufacturing is Key

Chris Klebanoff, MD, a CAR-T expert from Memorial Sloan Kettering Cancer Center (MSKCC) credits the TIL science first, but also the cohort of patients with advanced cervical cancer treated with the Iovance schema.

Efficacy aside for the moment, “Cell therapies will live and die by manufacturing,” says Klebanoff. “This lesson was painfully learned by Novartis, where, despite two indications for their CD19 program, including a first-in-class indication in pediatric ALL, they have really, really suffered relative to Kite/Gilead because of challenges with manufacturing.” Whereas Iovance, early on, devised a much more reliable, and a much faster turnaround cell manufacturing schema prior to larger scale investigations. “They can turn around cells in a little over three weeks.”

Returning to the efficacy of TIL, Klebanoff emphasized duration of response. “Like with checkpoints, cell-mediated CRs tend to be extraordinary durable. These CRs (n=3) are extremely meaningful.” As for the PRs, Klebanoff cautions that not enough time has passed to know if there will be an impact on overall survival.

TIL and Melanoma

The highly mutated tumors of melanoma are a favorite target of IO treatment. In an ASCO-reported trial of 66 advanced melanoma patients, results for the TIL drug, LN-144, administered as above, showed a cohort-wide reduction in tumor volume of 81%, including 2 CRs, 23 PRs, and 28 patients with stable disease (Figure 2). Strikingly, patients with PD-L1-negative status were among the responders. Furthermore, at a median follow up of 8.8 months, the median duration of response had yet to be reached.

 

A registration trial for LN-144 is now recruiting, and FDA approval thereafter seems fairly certain.

“Once FDA approval is achieved,” said study investigator, Amod Sarnaik, MD, H. Lee Moffitt Cancer Center, “There’s no reason why [TIL therapy] can’t be done first and, in my opinion, it makes sense to do this first and then PD-1 second. The reason for that is if you do PD-1 first, those patients tend to be worse off in terms of lack of reserve,” meaning, performance status. “So if you’re younger, in good health, if both treatments have the same response rates… why not have two bullets to kill the tumor instead of just one?”

Meso CAR, and CAR-T Costs

Progress in treating solid tumors with cell therapy is also being made in the arena of CAR-T cells—autologous T cells genetically engineered to recognize a specific target.

Prasad Adusumilli, MD, Memorial Sloan Kettering, presented his data for the efficacy of a CAR-T construct targeting mesothelin. “As a thoracic surgeon, my interest is to develop CAR-T therapy for pleural cancers,” of which there are 150,000 newly diagnosed patients in the U.S. every year, with an overall survival of 11 to 18 months. “The challenge is to move this technology into solid tumors,” which in this setting is a particularly high bar. “These typically are tumors with very low tumor mutational burden, and PD-L1 staining (meaning, not very immunogenic).”

Adusumilli hopes to overcome some of these challenges with administration of CAR-T cells directly into the pleura—assuring that the treatment can at least get to the target (one of many issues of using adoptive cell therapy in solid tumors).

For this phase I investigation, Adusumilli treated 27 patients with primary mesothelioma, or metastatic breast or lung cancers, with a single pleural administration of CAR-T cells under standard preconditioning protocols plus anti-PD-1.

Results are early but encouraging. First, the treatment was well tolerated—no cytokine release syndrome was observed. Second, though efficacy was modest, there was efficacy in this difficult-to-treat patient population: 3 CRs, 7 PRs, and 3 patients with stable disease, for an ORR of 63%.

“This study strongly supports pursuing CAR-T-cell therapy combined with PD-1 blocking strategy in solid tumors,” said Adusumilli, “We are transforming cold (non-responding) solid tumors to hot.”

Will it be cost-effective?

So, what is all this going to cost, and will it be cost effective? John Lin, MD, of Stanford University studied this question, and his ASCO data suggest that, at current price points, it is not going to be cost effective. Lin specifically looked at the two currently approved CAR-T products, Yescarta and Kymriah, in the setting of adult diffuse large B cell lymphoma (DLBCL). The price considered was $375k.

“Our conclusion was that CAR-T is probably not cost effective unless it meets our most optimistic long-term assumptions,” meaning, everyone in CR now stays that way, said Lin. “We only have data for CAR-T into two years so we don’t know what fraction of patients end up getting cured from CAR-T, so if you assume that most who are surviving at two years end up getting cured, CAR-T could be cost effective, if fewer that that are cured, then it’s not cost effective.”

Three notes: 1) This analysis would be very different in the pediatric setting, 2) For CAR-T in any setting late relapses are now being reported, and 3) Approved indications for CAR-T cells are expanding.

“Even at half the current cost, if CAR-T therapy became commonplace, these prices would be unsustainable,” Lin concluded.

Neoantigen Vaccines

Wrapping up this review is a presentation on a neoantigen vaccine by Moderna. Neoantigens are rarely shared between patients with the same type of cancer—a major challenge for cell therapies with targeting mechanisms—thus, a highly personalized approach is suggested, and is being investigated by multiple companies. Moderna’s entry in the race is mRNA-4157, a tailored-to-each patient vaccine encoding up to 20 neoantigens, with antigen selection based on the patient’s human leukocyte antigen (HLA) type and an analysis of the tumor mutanome, as accomplished by whole exome sequencing.

Once identified (there is a black box algorithm here, as there is with all neoantigen vaccine platforms) the RNA coding sequences for each antigen/peptide are incorporated in a nanoparticle. The preparation is then injected intramuscularly, where it is picked up by the dendritic cells of the immune system, where antigens are then expressed and processed for presentation (the last part being the mechanism whereby all vaccines against infectious disease work).

In a phase I, open label, multicenter dose escalation study, mRNA-4157 was administered, along with pembrolizumab (every 21 days for two cycles) to 20 patients with a variety of tumor types, including NSCLC, SCLS, bladder cancer, micro-satellite instability (MSI)-high patients, and patients with high tumor mutational burden.

Regarding safety, the observed grade 3 adverse events were consistent with those seen for anti-PD-1 monotherapy. As for efficacy, there was 1 CR (MSI-high patient), 5 PRs (two of which were PD-1 refractory patients at study entry, and one of which was a patient who discontinued pembrolizumab on study due to immune-related adverse events), and 6 patients with stable disease.

Of note, 10 patients remain on treatment with ongoing clinical benefit as of May 2019. Of further note, Howard, “Skip” Burris, of Sarah Cannon Research Institute and current president of ASCO was the PI on the trial.

In an interview, Burris told OBR as molecular testing increases, it becomes clear that these patients have markedly different profiles. “The idea of actually designing a vaccine that would be individualized to the patient’s profile is very attractive. I mean, years ago we were involved in the THERATOPE program, a one size fits all breast cancer vaccine, but it just didn’t work out.”

“We also did trials with CEA [carcinoembryonic antigen]-directed, one size fits all colon cancer vaccine [that also crashed], so this idea of a personalized vaccine approach to match what we learned from molecular profiling is unique.”

And the data showing that it might actually work certainly adds to the appeal.

 

Post a Comment

OBR Archives

To view previous issues of OBR green you can visit our archives. The entire library of OBR green articles is searchable.