Skip to main content
Premium Trial:

Request an Annual Quote

A2 Biotherapeutics Enrolling Patients in Clinical Study of New CAR-T Therapy Using Tempus NGS Panel


This story has been updated from a previous version posted Feb. 8 to clarify that the xt-Onco NGS assay was not developed as part of this collaboration but was used in it to detect hard-to-find mutations.

NEW YORK – As A2 Biotherapeutics continues to advance a new type of CAR-T cancer therapy that targets loss of heterozygosity (LOH), it is partnering with Tempus to develop a next-generation sequencing companion diagnostic panel to identify patients most likely to benefit from this treatment.

Following several proof-of-concept studies, A2 is now enrolling participants in an observational study to identify individuals potentially eligible for the autologous CAR-T therapy, based on The company's so-called Tmod platform, and to store their T cells for treatment in follow-on interventional studies.

LOH describes the loss of one normal copy of a gene or group of genes and is a common event in oncogenesis.

The Tmod CAR-T platform is unique in its "AND NOT" logic gating strategy. It targets cells that have one marker and not another, through the combined action of activator and blocker elements. Each element is an epitope that recognizes another cell surface protein, one of which is found on an individual's tumor cells (the activator) and one of which has been lost on tumor cells but remains present on normal cells (the blocker).

When both the activator and blocker elements find their binding partners on a cell, the CAR-T does not receive a signal to kill that cell. When the activator is present and the blocker is not, the CAR-T is activated and kills the cell.

Whereas most CAR-T therapies will seek a compromise between the CAR-T's cancer cell specificity and off-target cytotoxicity, "we do the opposite," Scott Foraker, president and CEO of A2, said in an interview. "We actually tune up our activators for maximum cell killing, because we have the benefit of the blocker protecting the normal cell."

A proof-of-concept study published in Molecular Immunology in 2020 showed that this method worked accurately both in vitro and in a mouse xenograft model, targeting CD19 as the activator and the HLA A*02 locus as the blocker. This allele is most prevalent in the North American population and undergoes LOH in cancers at a relatively high estimated frequency of 13 percent. The study further established that the HLA LOH frequencies in a range of gastrointestinal cancers are high enough to make HLA LOH a trustworthy discriminator of tumor versus non-tumor cells.

Earlier this month, A2 published another study showing that the platform worked when targeting mesothelin as the activator and again, HLA A*02 as the blocker, also in vitro and in a mouse xenograft model.

Mesothelin is a well-known tumor-associated antigen but a risky therapeutic target, as it is expressed both in lung cancer and other solid tumors, as well as on normal healthy cells. The study showed that mesothelin-specific Tmod CAR-T cells killed tumor cells with a sensitivity comparable to that of a benchmark mesothelin-specific CAR-T, but without the associated cytotoxicity.

The company is also exploring a range of other activators, such as the carcinoembryonic antigen (CEA), a solid tumor biomarker often associated with colorectal cancer.

A2 is also looking beyond HLA as a blocker for targeting LOH, although the company declined to go into specifics.

"We think the technology can be expanded far beyond our initial study of HLA A*02 right into other HLAs and then beyond…into new target classes," Foraker said.

A2 is currently recruiting participants for the observational BASECAMP-1 clinical study, aimed at identifying and banking T cells for individuals potentially eligible for autologous Tmod CAR-T therapy.

Foraker expects to begin phase 1 basket studies — which enroll patients with a certain genetic mutation in common regardless of their site of origin — for mesothelin and CEA-targeted Tmod next year.

A2 has worked with Tempus to identify patients with LOH who are most likely to benefit from Tmod CAR-T therapy. A2 tested the ability to identify potentially difficult-to-find mutations via Tempus' new xt-Onco NGS assay, for which the company recently submitted a premarket approval application to the US Food and Drug Administration.

The submission for the 684-gene NGS-based lab-developed test (LDT) includes companion diagnostic claims and the detection of single nucleotide variants (SNVs), insertions and deletions of under 100 base pairs, copy number variants, chromosomal rearrangements, and microsatellite instability (MSI) status, all using DNA isolated from formalin-fixed paraffin embedded tumor tissue specimens, as well as from matched normal specimens.

Complementary RNA sequencing further enables the unbiased detection of additional clinically validated gene fusions.

Although other assays exist that can identify LOH and are potentially cheaper than NGS assays, such as PCR and FISH, both A2 and Tempus say that these fail to discriminate targets sufficiently.

"NGS is necessary," Lauren Silvis, senior vice president of external affairs at Tempus, said via email. "We are looking to identify patients that might be difficult to find, and our test can be utilized across cancer patients to help identify those who might benefit [from Tmod CAR-T]."

William Go, chief medical officer at A2, explained that Tempus' broad solid tumor NGS coverage made it an ideal partner, and that its xt-Onco test is the "one test to rule them all," in terms of identifying targetable LOH activator-blocker combinations.

"This is the next generation of next-gen sequencing," he said. The combined germline and somatic information, combined with the assay's other features, allows one to "really discriminate the tumor versus normal [tissue]."

A2 plans to manufacture its Tmod therapies at its own fully integrated cGMP manufacturing facility.

"We already had a pre-facility FDA meeting and we already are now doing engineering runs of multiple different programs, both auto and allo, in one GMP suite," Go said.

A2 has so far raised $200 million in funding, providing a cash runway expected to last at least the next two years.

"So we're not at all dependent on near-term market gyrations," Foraker said.

A2 and Tempus plan to continue their collaboration for the foreseeable future.

"A2 Bio has a robust pipeline with autologous and allogeneic therapies," Silvis said, "which will require this continued partnership to support the development of further diagnostic testing for future programs."