CAR-T CELLS FOR MESOTHELIOMA

Caligiuri indicated that it was too early to tell whether the responses were due to CAR T cells themselves. At the press conference, Adusumilli pointed out that all approaches that used immune blockade alone in the treatment of mesothelioma have shown poor responses, which were not durable. He noted that the persistence of CAR T cells and the decline in mesothelin markers suggested that the response was due to CAR T cells, which worked in concert with immune checkpoint blockade.

The Novelty of the Approach

The first novelty in this approach was in the manner in which toxicity concerns were addressed, Adusumilli commented. His team engineered an IcasM28z safety switch into the mesothelin-targeted CAR T cells. IcasM28z is a prop-apoptotic suicide switch that can be activated by an intravenous medication to kill all CAR T cells in the patient should an unexpected toxicity arise.

The second novelty was in determining that it was best to inject the CAR T cells directly into the pleural cavity. "We had observed that when we injected the CAR T cells into the blood, they lodged in the lungs for 4 to 5 days before making their way into the tumors at low levels," Adusumilli explained. "By injecting the cells into the pleural cavity, the CAR T cells multiply into millions and initially bypass the lungs," he said.

A third novel feature of their approach was in linking CAR T cells with checkpoint blockade agents. "When we first used these cells in mice, we noticed that CAR T cells get 'exhausted' in the presence of overwhelming tumor burden," Adusumilli commented. "However, we could rescue these cells with immune checkpoint blockade that reactivates the cells," he added.

Malignant pleural disease from primary malignant pleural mesothelioma or secondary metastatic disease, such as lung and breast cancer, affects more than 150,000 patients in the United States alone.

"If this approach is successful, 2 million patients with mesothelin-expressing solid tumors per year in the United States would be eligible for this treatment," Adusumilli said.

Study Details

Adusumilli and colleagues treated 21 patients who had malignant pleural disease with second-generation, CD28-costimulated fully human mesothelin CAR T cells with the Icaspase-9 safety gene (IcasM28z). Of the 21 patients who were treated, 19 had malignant pleural mesothelioma, one had lung cancer, and one had breast cancer.

On-target/off-tumor toxicity was determined from laboratory measurements of troponin elevation as well as from CT/positron-emission tomography (CT/PET) scans or echocardiograms for pericardial effusion and ascites and routine clinical evaluation.

Patients were given a single dose of the CAR T cells (3 x 10⁵ – 1 x 10⁷ cells/kg), which was administered intrapleurally by a pleural catheter or through an interventional radiology procedure. Eighteen patients received a conditioning regimen of cyclophosphamide. Twelve patients were given the CAR T cells via an interventional radiology procedure.

CAR T-cell persistence was reported in 13 patients. In these patients, the CAR T cells persisted from day 2 to 42 weeks. There were associated responses; a decrease in serum SMRP (soluble MSLN-related peptide) level; and tumor regression, as seen from imaging scans.

After it was established that there was no evidence of CAR T-cell–related toxicity (ie, no neurotoxicity, no cytokine release syndrome, and no on-target/off-tumor toxicity), patients received a checkpoint blockade agent — an anti-PD1 antibody — off protocol (range: 6 – 17 weeks after CAR T-cell infusion).

Fifteen patients received anti-PD1 therapy (range: 1 – 21 cycles) off protocol with no reported toxicity. The best response was seen in 19 patients with malignant pleural mesothelioma; 13 of these patients received anti-PD1 therapy. Two patients had a complete response, as seen from PET scan; the responses were ongoing (60 weeks in one patient and 32 weeks in the second). Six patients had a partial response, and four had stable disease.

In the subgroup of 11 patients who received cyclophosphamide, CAR T cells, and at least three doses of anti-PD1 therapy with a minimum of 3 months of follow-up, the response rate was 72%.

What's Next for Mesothelin-Targeted CAR T Cells?

Further studies will evaluate the efficacy of this approach with escalating doses of CAR T cells. In addition, Adusumilli and colleagues will receive approval from the US Food and Drug Administration to deliver immune checkpoint blockade earlier, at week 3.

The team is also working on a strategy to engineer CAR T cells with a decoy receptor, which is a PD1 dominant-negative receptor. "This approach has worked well in mice and is an anti-PD1 strategy that takes off the signaling component. There will be no need for immune checkpoint blockade with this approach," he said.

MSKCC has licensed mesothelin CARs to Atara Biotherapeutics, has received license fees, and may receive royalties under the license. Adusumilli and Sadelain are inventors and will receive a share of the license income. Adusumilli's and Sadelain's laboratories receive sponsored research funding from Atara Biotherapeutics. Caligiuri is on the advisory board of CytoSense Therapeutics and Cellular Biomedine Group.

American Association for Cancer Research (AACR) 2019: Abstract CT036, presented March 31, 2019.