Skip to main content
Premium Trial:

Request an Annual Quote

Carisma Therapeutics Taps Into Innate Immune System to Extend CAR Therapy to Solid Tumors


NEW YORK – Carisma Therapeutics last month infused the first patient with a HER2-overexpressing solid tumor with its CAR macrophage therapy CT-0508, which may have been the first time any patient has received engineered CAR macrophages, according to the Philadelphia-headquartered drugmaker.

The firm is betting that if this development program is successful, it will usher in a new era of cell therapies for solid tumors.

During the American Association for Cancer Research's annual meeting, Carisma provided an in-depth overview of this recently launched first-in-human trial and presented its plans for speeding up the manufacturing process of its cell therapies to a single-day turnaround time. The firm also presented preclinical research on its HER2-tartgeting CAR-macrophage (CAR-M) product, shedding further light on how the cells interact with the tumor microenvironment.

Central to all these presentations is the overarching notion that unlike CAR T cells, CAR-M therapy can effectively treat solid tumors by overcoming the mechanistic heterogeneity of cancer cells that allows them to bypass the immune response evoked by single-antigen therapies. Researchers are hopeful that CAR-M therapies, unlike CAR T cells, may also work in cold tumor environments that evade T cells. By restricting the CAR macrophages to a pro-inflammatory anti-tumor "M1" phenotype, Carisma believes its treatment can overcome these barriers and recruit immune cells to historically impenetrable tumor environments.

From dissertation to clinical-stage company

Having just entered the clinic, Carisma's CT-0508 still has a long way to go in terms of demonstrating its safety and efficacy in HER2-overexpressing solid cancers. Still, the fact that the company was able to successfully harvest innate immune cells from a cancer patient, select and engineer the cells into CAR macrophages targeting HER2, then cryopreserve and reinfuse the treatment into the patient, represents a massive leap from the treatment's conceptual beginnings in 2014.

Carisma's Co-founder and Senior VP Michael Klichinsky laid the foundation for the company in his PhD dissertation at the University of Pennsylvania. "CAR T cells have been tremendously successful in blood cancers … but in solid tumors, the responses in clinical trials have been essentially zero across the board," Klichinsky said. "And the idea that I had with Saar Gill [Carisma's other cofounder] was that there was perhaps another immune cell better suited for solid tumors."

That immune cell type wound up being macrophages, Klichinsky explained, because macrophages are innate immune cells that are actively recruited and can effectively "flip the environment on its head," converting cold, immunosuppressive tumor environments into warm, pro-inflammatory tumor environments. Macrophages are also antigen-presenting immune cells that can "not only kill based on the target you tell them to kill but can then amplify their own effect and trigger the adaptive immune response to go after patient-specific mutations," he added.

After landing on the macrophage as the immune cell type they would pursue, Klichinsky and his collaborators faced a challenge: how would they engineer these unruly cells? "At the time, a key barrier in the field was an inability to engineer these [macrophage] cells," he explained. "They're very difficult to work with, and the methods used to make CAR T cells or CAR NK cells do not work on macrophages."

Specifically, the lentiviral vector used to engineer most CAR T cells doesn't work with macrophages. Knowing this, then-PhD candidate Klichinsky spent over a year testing out different viral vectors, ultimately landing on the adenovirus vector Ad5f35. This viral vector has a "genetic trick" in the fiber that gives it the ability to infect cells that express the protein CD46, like macrophages and monocytes. "The virus is able to infect these cells with great efficiency, so we're getting CAR expression in 90 percent of human monocytes and macrophages by using this vector," Klichinsky said.

Finding a vector that enables macrophage engineering opened up possibilities in terms of targeting solid cancers. "We demonstrated that once we put a CAR into a macrophage, we can tell it what to eat through phagocytosis [and] what to kill," he recalled.

The preclinical research in Klichinsky's dissertation demonstrated the therapeutic potential of these cells to such an extent that in 2016 Klichinsky and Gill wanted to actually use them to develop drugs. They launched Carisma, which five years later has grown to a company of more than 40 employees. The firm has also completed multiple rounds of financing, closing a series B round in early March with proceeds totaling $59 million, bringing the company's total capital raised close to $121 million.

Trial design, target selection

The recently launched Phase I basket trial evaluating CT-0508 will enroll roughly 18 patients with various HER2-positive solid tumors at the University of Pennsylvania and the University of North Carolina. According to Deborah Barton, Carisma's chief medical officer, the trial will use immunohistochemistry assays to select patients with HER2-overexpressing tumors for the trial, including those with breast and esophageal cancers, but also patients with other HER2-positive tumor types lacking treatment options.

For the indication of its first CAR-M product, the company settled on HER2-positive tumors because it is a well-established and well-studied setting. "This is an entirely new approach, and we didn't want to simultaneously test a new platform and a new, unproven target," Klichinsky explained. Additionally, the fact that HER2 is overexpressed on a variety of solid tumors was a draw, since it allows the company to cast a wide net in the Phase I study and evaluate CT-0508's activity in a variety of tumor types.

Safety, noted Klichinsky and Barton, is another advantage of the CAR-M approach. As the trial continues, Barton doesn't expect that patients will experience adverse effects common to CAR T-cell therapies, such as cytokine release syndrome or neurotoxicity.

"Macrophages have several differences versus T cells," she said. "For one, they don't expand … and the macrophages leave the bloodstream very quickly, [so] the cytokines they produce, they produce in tissues and not in the blood." Additionally, unlike CAR T-cell therapies, patients don't need lymphodepleting chemotherapy prior to treatment with the engineered macrophages, which may also cut down on toxicities. Still, Carisma's clinical teams have implemented safety measures to manage these toxicities should they occur.

Beyond evaluating safety and tolerability of the product, the Phase I trial will also assess the feasibility of manufacturing CT-0508, as well as initial efficacy endpoints like objective response rate and progression-free survival. In terms of efficacy, even though CAR-M cells do not have the same adaptive immune system effect that has enabled CAR T cells to evoke sustained, years-long responses in some blood cancer patients, Barton highlighted other characteristics of CAR-M therapies that may lead to durable benefits.

Pro-inflammatory macrophages, for example, have the added advantage of attracting the adaptive immune system to the tumor environment. "[The CAR-Ms] will be able to phagocytize the tumor cells then trigger the adaptive response that should be long-lasting," Barton said.

Toward single-day manufacturing

Currently, the process of manufacturing CT-0508 takes about a week and involves partnerships with disparately located manufacturing sites. The vectors are assembled at WuXi Advanced Therapeutics in Philadelphia, and the primary cell manufacturing takes place at Miltenyi Biotec's facility in San Jose, California. According to Klichinsky, once other steps are factored in like transporting the cells and testing the manufactured product before infusion into the patient, the full vein-to-vein time from leukapheresis to infusion can take up to three weeks.

Similar to manufacturing CAR T-cell therapies, the CAR-M process involves engineering the cells using the viral vector encoding a CAR transgene. With CAR-M cells, however, there are added steps, including selecting CD14-positive monocytes and differentiating them to macrophages. As the Carisma team laid out in a poster at AACR, this part of the process may be unnecessary if the company can directly engineer CD14-positive monocytes themselves.

"We essentially asked the question, 'What if we just take the monocytes and engineer them directly and make CAR monocytes?'" Klichinsky said. "The CAR monocytes will just differentiate naturally into macrophages in the body as monocytes normally do … [Then,] can we skip that entire monocyte-to-macrophage differentiation process and just engineer them as monocytes directly?"

Klichinsky said the company has solved this question by purifying the monocytes from blood and engineering them with the viral vector immediately. Carisma has demonstrated preclinically the feasibility of this so-called "CAR-mono" approach and in an AACR presentation said it could shorten the manufacturing process from over a week to just 12 hours. "This would be quite significant" in terms of treating patients faster and lowering the cost of goods, Klichinsky said.

The shortened, monocyte-engineered manufacturing process has so far been tried in a HER2-positive ovarian cancer xenograft model but not yet in humans. Before this approach can enter the clinic, Klichinsky said further testing will be needed in animal models and additional process development work must be done to show that it is feasible to engineer and scale CAR-mono cells.

After getting a handle on CT-0508's activity in HER2-positive solid tumors, Carisma will set its sights on other targets, such as mesothelin in solid tumors and PSMA in prostate cancer, which both are indications in preclinical development.

"This is a brand-new field, CAR macrophages," said Klichinsky, though he acknowledged there are still many uncertainties around making these cells, identifying best responder populations, and selecting ideal targets. "We think we have a good sense of these things, but we're approaching this scientifically and letting the data drive the direction of the company."