Skip to Main Content

A novel approach to CAR-T cancer therapy promises to upend what has become a truism in medicine: that the treatment’s dramatic effect on tumors comes with serious risks to patient safety.

In a small clinical trial, patients treated with a slightly modified version of an approved CAR-T saw benefits on par with previous studies but with none of the hallmark side effects that can leave some patients hospitalized and in need of costly extra treatments.


The study enrolled just 25 patients in China, according to data published Monday. But if the results can be replicated, it suggests that a little molecular tinkering could make CAR-T safer and more widely available, experts said.

“This looks very promising,” said Jill O’Donnell-Tormey, CEO of the nonprofit Cancer Research Institute. “Obviously it’s early days, but the response they’ve had in the 25 patients they’ve treated is impressive.”

CAR-T therapies are crafted by extracting a patient’s own immune cells, genetically engineering them to target cancers, and then equipping them with compounds meant to recruit the body’s natural defenses to join the fight. To craft a safer treatment, scientists took aim at the last step in that process.


They began with Novartis’s Kymriah, approved to treat a pair of blood cancers, and then made analogs of their own, each one stitched together with a slightly different sequence of amino acids. When they tested these variants in mice, they made a striking observation: One of the modified CAR-Ts was killing cancer cells without triggering a feverish immune response or causing inflammation in the brain, the two most common serious side effects of cell therapy.

And it held up in human testing. In the study, published in Nature Medicine, the remixed Kymriah led to no serious cases of cytokine release syndrome, an immunological flareup common in CAR-T, and no incidents of neurotoxicity. By contrast, more than half the patients in Novartis’ published studies suffered from cytokine release, and about a quarter had neurological problems.

“That, to us, was a very big surprise,” said Dr. Si-Yi Chen, a professor of immunology at the University of Southern California and lead author of the paper.

It was a surprise to Chen’s colleagues, too.

The modified CAR-T appeared to hit an immunological sweet spot, recruiting enough cytokines to have an effect on cancer without stirring up any trouble. But it’s not clear why. It could be that approved CAR-Ts, like Kymriah and Gilead Sciences’ Yescarta, are simply too strong and that a comparatively weaker therapy can do the same work with less risk. Or it could be random chance.

“I would look at this with a bit of caution, or cautious optimism,” said Dr. Loretta Nastoupil, director of the lymphoma department at MD Anderson Cancer Center in Houston. “Understanding the mechanisms of what’s driving this efficacy will be critical.”

Beyond the underlying science, there’s the question of durability. Approved CAR-Ts often lead to long-lasting remissions. Chen’s approach looks safer in the short term, but it remains to be seen whether the effects can endure, said Dr. Michel Sadelain, an immunologist at Memorial Sloan Kettering Cancer Center.

“The concern would be, well, if you weaken the CAR, that’s great if you diminish the cytokine release, but might you diminish the therapeutic effect?” Sadelain said. “That’s the big question mark here. And only time will tell.”

Those questions aside, the promise of a safer CAR-T could dramatically broaden access to a treatment that is now available only at major cancer centers. Dealing with the treatment’s side effects often requires specialized care and expertise not available at community hospitals, limiting the number of patients who can get treated.

And then there’s the cost. The sticker price of CAR-T treatment is upward of $370,000 for a treatment, but that doesn’t include the cost of hospitalization and drugs used to tamp down immune response. In the most severe cases, the end price often exceeds $1 million, said Avery Posey, an immunotherapy researcher at the University of Pennsylvania.

That combination of immunological side effects and neurotoxicity “is what the residents at Penn call ‘CAR-Tastrophy,’” Posey said.

The Nature Medicine paper suggests a future free of those problems.

“That is one of the most discussed aspects of CAR-T cell therapy: If you can minimize toxicity, you may be able to minimize costs, and that might improve access,” Nastoupil said.

But first the purportedly safer CAR-T needs to be tested in more patients, she said, and the researchers need to demonstrate that it leads to the kind of lasting remissions seen with Kymriah and Yescarta.

Chen and his team want to do just that, mapping out a Phase 2 trial that would dose more patients and follow them for more time.

It’s unclear who will finance that effort. Chen struggled to finance the first clinical trial, he said, told time and again that his CAR-T innovation was too incremental to be worthwhile. Marino Biotechnology, a private Chinese firm, ended up footing the bill, but the company’s future investment hangs on the key question of whether Chen’s modified version of Kymriah is actually patentable.

  • Yes, why can’t the investigational new CAR-T cell cancer chemotherapy be offered freely under compassionate use while the novel therapy is being studied ? Are we in phase 4 or post-marketing surveillance stages, with volunteer patients as essential study subjects, really acting as courageous human guinea pigs?

  • Come on, people! If the data are available for review and really good, crowdfunding or government funding could help here, and companies could take a piece if they were willing to take a risk. People are dying every day who could benefit from a breakthrough like this! Those people and their loved ones don’t care if it’s patentable, and neither should Novartis!!!

Comments are closed.