It’s a big week for oncology and immunotherapy, as an FDA panel convenes on Wednesday to interrogate the first CAR-T therapy vying for approval.

Developed by Novartis, the five-step approach seeks to harness and weaponize the patient’s own immune system to fight off late-stage blood cancers. Kite Pharma has its own version awaiting FDA review and several others are moving through clinical trials.

With their almost science fiction-like complexity and remarkable effects, Chimeric Antigen Receptor-engineered T-cells (CAR-Ts) have understandably captured the field’s attention. But underneath the hype and excitement, all the different programs rely on a much more basic and familiar aspect of cell biology: the protein/receptor target.

Whether it’s a cancer vaccine, an antibody-drug conjugate (ADC), or an entirely new platform; there has to be a way for the therapy and/or the immune system to differentiate between cancerous and healthy cells.

It’s a common denominator throughout all immuno-oncology programs and as a result, it can be a window into the evolution and expansion of the field.

All or nothing
According to a recent report by the trade group PhRMA, there are 240 cancer immunotherapies in the pipeline. For each candidate, R&D teams have to ask the same fundamental question: How can we drive or enable an immune response?

The lead programs of both Novartis and Kite are actively programming T-cells to seek out CD19, a protein found on the surface of mature B-cells that drive the blood cancers they aim to treat. Two other CAR-T companies, Bluebird Bio and Nanjing Legend Biotech, are targeting a receptor known as BCMA.

When it comes to solid tumors, it gets a bit harder. The tissue that the cancers arose from (pancreatic, lung, colon) has to be protected. It becomes more of a trade-off; how widely is a protein/receptor expressed on cancer cells and how often is it present in healthy tissue? No target used today is perfect, so there are inevitably side effects and a percentage of the tumor cells may evade the drug and/or the immune system.

Attend MedCity CONVERGE, a cancer innovation conference,  to hear from diagnostics, Big Data, gene editing, and other precision oncology experts. Use promo code MCNPOST to save $50. Register now.

Whichever cancer is being targeted, the key to even greater efficacy may lie with earlier interventions.

Given the complexity and the potentially life-threatening side effects, CAR-T trials have focused on patients that have exhausted all other options. For example, participants in Bluebird’s lead Phase 1 trial had taken between three and 14 therapies each, with at least one autologous stem cell transplant.

In a June interview announcing the data, “Chief Bluebird” (CEO) Nick Leschly told MedCity News that if toxicities are adequately managed, “you can also contemplate, how do you treat these patients earlier? Because if you treat them earlier, you might get even better responses with even more durability.”

One of the key benefits of an immunotherapy approach is that it can lead to the formation of memory T-cells. These stay in the system for years and can rapidly multiply and destroy any recurrence of the tumor. It’s not like a drug that is eventually cleared from the body.

Triggering a response
Another promising new area of research is targeting neoantigens. These are mutations that arise “de novo” in a given cancer — they’re not otherwise found in the human genome.

As an immunotherapy target, they offer two major benefits: They’re foreign to the immune system and they’re not found in healthy tissue.

By comparison, traditional lung cancer targets such as ALK or EGFR have been present in the body since early development. They may be overexpressed in cancer cells, but the immune system has over time learned to tolerate them as “self.” That’s not a good platform for triggering a T-cell attack and it raises the likelihood of off-target effects. 

While promising, the use of neoantigens as a cancer bullseye is still in its infancy. In fact, just a handful have made it to clinical trials, noted Merck’s SVP, Head of Global Clinical Development, and Chief Medical Officer Roy Baynes.

A lot of the challenge for scientists looking to trigger an immune response against neoantigens is finding them in the first place. It’s like coordinating a search for a wanted criminal: You have to have a picture or a description to show the immune system what to look for. Otherwise, the tumor gets lost in the cellular crowds.

“There’s a lot of work going on,” Baynes said. “People are coming at this from a number of different angles, either proteins, peptide, miRNA, DNA. The idea being, can we identify any of these new antigens?”

To that end, Merck has a collaboration underway with Moderna Therapeutics. The partnership’s aim is to identify unique mutations in a given cancer, Baynes explained, as the first step in a cancer vaccine approach that would excite a very specific immune response.

“But there are a number of other approaches ongoing as well,” he noted.

In fact, just last week promising data on a cancer vaccine was published in Nature. It involved just six melanoma patients, but it was a notable first step towards human efficacy. All six patients survived without their primary tumor returning and after two years, their T-cells still showed signs of activation against the melanoma.

Removing the brakes
In some instances, there may be no need to excite an immune response. All that needs to be done is to expose the tumor and its mutations.

Merck has found huge success with Keytruda (pembrolizumab), part of a class of drugs that disrupt PD-1/PD-L1. Known as immune checkpoints, some cancers use these mutations as a way to hide from the immune system.

Two other targets, CTLA-4 and OX40, work in a similar way by downregulating the immune response. All told, the PhRMA report identified 45 checkpoint inhibitors in development, underscoring their pivotal role in today’s immunotherapy landscape.

The idea is that when the tumor’s cloak is removed, the immune system can truly interrogate the cells. If the mutational burden (the total number of novel mutations) is high, then there is a good chance that it will be able to recognize the cancerous cells as a threat that should be destroyed.

Alternatively, if the tumor burden is low, there may be nothing to excite the immune system. A second or third drug may be needed to boost the response. To that end, there are currently more than 500 clinical trials for Keytruda as a mono or combination therapy registered on

The scope of Keytruda’s potential application highlights how much we know and how little we know.

On the one hand, scientists are rapidly uncovering new, actionable information about the biology of cancer. This was exemplified by the recent landmark FDA approval of Keytruda for use in a range of solid tumors that carry certain biomarkers. More specifically, it was indicated for cancers that are microsatellite instability-high (MSI-H) or mismatch repair deficient (dMMR). Both are abnormalities that affect the proper repair of DNA inside the cell (aka mismatch repair genes).

It was the first “tissue agnostic” ruling ever made in this field and as Baynes noted, it illustrates the potential for targeting novel mutations.

“The notion here is that when you have a defect of DNA repair, one is more likely to see mutations in the cancer cell and these mutations should give rise to neoantigens a percentage of the time,” Baynes said. “So that’s sort of the underlying thesis as to why the MSI-high population is really so responsive to checkpoint inhibition modulation.”

Conversely, we’re still largely treating cancers with a piecemeal approach. As John Quackenbush, a professor at the Dana-Farber Cancer Institute, director of its Center for Cancer Computational Biology, and a keynote speaker at the upcoming MedCity CONVERGE conference in Philadelphia noted, tumors aren’t a series of individual mutations, they’re a whole that has evolved to evade the immune system.

“What we’ve done in a lot of ways in implementing precision medicine, is we’ve taken a very simplistic view of how biological systems functional,” Quackenbush said in a recent phone interview. “And the fact that so many people fail, or even the fact that people who shouldn’t respond, based on what we know, actually do respond at some rate to these therapies, really suggest that if we want to move the field forward, we’ve got to take a much more integrated view of the complexity of biological systems.”

The good news is that tumors are being sequenced and databases are being built. We can learn in real terms what worked and how patients responded to therapies. When outcomes don’t make sense, machine learning can be used to make sense of it.

Over time, this will lead to more biomarkers and more targeted drugs.

Work to be done
One of the first immunotherapies approved, Yervoy (ipilimumab) didn’t work across the board. But after 10 years, 17 percent of patients with late-stage melanoma were still alive. As noted by Medscape:

“In the pre-ipilimumab era, patients with advanced metastatic melanoma were treated with chemotherapy and interferon, and survival was measured in months — on average, 10 to 11 months, with a range of 6 to 18 months.” 

Further gains have been made with checkpoint inhibitors and other personalized medicines. Melanoma, in a nutshell, has been one of the success stories.

Other cancers have hardly budged. Pancreatic cancer remains incurable: After five years, the relative survival rate for patients is just seven percent, according to the American Cancer Society. The tumors are deeply seated and often surround themselves in a tough, fibrous shell that prevents T-cells from accessing it. Brain cancers are also extremely difficult to treat and the wider organ is by design, impervious to immune cells. Other approaches will be needed.

The war on cancer must push on.

Photo: jxfzsy, Getty Images