ORLANDO, Fla. — There was some drama at the American Association for Cancer Research annual meeting: stocks crashed, companies debuted new compounds and trial results, and a blood cancer mystery inched closer to being solved. If you missed the conference here this past week, here are some highlights.
Moderna and Merck showed the first randomized clinical evidence that their melanoma vaccine mRNA-4157, in combination with pembrolizumab, can reduce the risk of relapse for patients, compared to pembrolizumab alone.
A new company called Palleon Pharmaceuticals also showed promising phase 1 data on a compound based on some of the Nobel Prize-winning work of Stanford biochemist Carolyn Bertozzi.
Relay Therapeutics‘ stock tanked after the company presented early data on their PI3K inhibitor. Some clinicians, though, still found the Phase 1 data encouraging.
Stanford pathology researcher Siddhartha Jaiswal presented work that put scientists a step closer to uncovering the drivers of CHIP, a condition that can give rise to leukemia and is associated with several health risks.
STAT also highlighted other research each day of the meeting in our newsletter “AACR in 30 Seconds.” Below are some of those notes.
Allogeneic CAR-T cells get edited and show safety
The use of CAR-T cells from healthy donors, or allo CAR-T, has long struggled with intrinsic challenges — notably the host’s immune system rejecting the donated CAR-T cells or vice versa, the CAR-T cells attacking the patient’s healthy cells in graft versus host disease. At the AACR annual meeting this year, two abstracts showed new data outlining progress in allo CAR-T products edited for better safety, stronger resistance, and increased potency.
One is from Allogene Therapeutics, which presented Phase 1 data on ALLO-316. This is a CAR-T cell targeting CD70 for renal cell carcinoma. As of March 2023, 18 patients had been evaluated for efficacy, 10 of whom had CD70 positive tumors. The overall response rate was 17%, although among the patients with CD70 positive tumors, the response rate was 30% and the disease control rate was 100%. The additional gene editing in ALLO-316, knocking out a key gene involved in graft vs host disease, seemed to pay off so far — no such disease was observed in this study.
CRISPR Therapeutics also showed preclinical results on two allo CAR-T cell therapies, CTX112 and CTX131, which are modified using CRISPR/Cas9 gene editing that hypothetically should improve the cell’s anti-tumor activity and persistence. Altogether, the work shows how the field is moving forward, said Sidi Chen, a geneticist at the Yale School of Medicine who did not work on the abstracts.
A pan-RAF inhibitor shows promise, but no wow from investors
BRAF mutations are oncogenic driver mutations found in many solid tumors including colorectal, pancreatic, and lung cancer, and come in three general varieties: Class 1, 2, and 3. Currently, there’s only drugs for BRAF class 1 mutations, but Kinnate Biopharma thinks it has a candidate that can hit all three classes. The data presented at AACR, however, failed to impress investors.
The Phase 1 trial results showed the compound, called exarafenib, is more able to enter the blood and interact with tumor cells than other BRAF inhibitors, said Nima Farzan, Kinnate’s CEO. Exarafenib also showed some responses in patients, notably in 1 of 3 patients with BRAF Class 2 mutations. Class 2 and 3 mutations make up over half of all patients with BRAF mutations. Across all 34 patients in the trial, 6 had partial responses.
“Really exciting monotherapy activity,” Farzan said.
The market, apparently, disagreed. When Kinnate’s abstract came online, its share price dropped 20%.
Guidelines on testing for BRCA or Lynch syndrome miss 40% of mutation carriers
People who carry one of the BRCA mutations in HBOC or hereditary breast and ovarian cancer syndrome, or one of the mismatch repair genes implicated in Lynch syndrome have a lifetime cancer risk of 80 to 90%. “It’s exceedingly likely you will have cancer,” said Niloy Samadder, a cancer researcher from the Mayo Clinic, making it extremely important for carriers of these mutations to know they have it and get the proper screening — like breast cancer screening for BRCA mutations and colonoscopy for Lynch syndrome.
But in an analysis Samadder presented at the annual meeting, many people who carry one of these mutations would be missed by current NCCN guidelines on who should be screened. He and his colleagues conducted whole exome sequencing on over 44,000 patients in the TAPESTRY study at the Mayo Clinic and came across 550 people with either a BRCA mutation or one of the mismatch repair gene mutations in Lynch snydrome. Of those, 40% did not meet the NCCN criteria for genetic screening. “This ranged between 56% of Lynch syndrome and 32% of HBOC patients,” Samadder said.
In addition, the criteria missed 50% of underrepresented minorities with HBOC or Lynch Syndrome, compared to 32% of white patients.
A deep learning model to say help guide who needs chemo after surgery
The primary treatment for endometrial cancer is surgery, after which adjuvant chemotherapy can help stop distant recurrence. But chemotherapy, of course, comes with unpleasant toxicities — and some patients may not need it or need it as intensely. Sarah Fremond, a graduate student in pathology at Leiden University Medical Center, presented work on a deep learning model that sorts patients into high or low risk for recurrence based on their cancer stage and digitized histopathological slides. The model, called HECTOR, was able to outperform an existing model called Baseline Cox.
To make the model, Fremond used a dataset from completed randomized trials called the PORTEC 1, 2, and 3 trials that tested adjuvant chemotherapy in endometrial cancer patients. This included 1761 patients that had a digitized histopathological slide, and an average eight years of follow up. HECTOR trained on a portion of these data to learn how to predict which endometrial cancer patients were more likely to get recurrence after surgery. Then, she tested the model in the remaining patients HECTOR had not seen.
“What we have built could theoretically be used for other cancer types,” Fremond said. “These studies are retrospective, so we need to validate HECTOR on prospective studies.”
A weak T cell engager might not make a bad bispecific
Regeneron had Phase 1/2 data presented at AACR that showed a bispecific antibody for multiple myeloma might be able to lower toxicity without sacrificing efficacy. The antibody, called REGN5459, works by binding the protein BCMA on myeloma cells. But it only loosely engages the T cell, using a low-affinity CD3 arm.
Approved bispecific antibodies like Janssen’s teclistamab for multiple myeloma have given patients a new, effective option in recent years, but it can also come with some heavy toxicity. By dialing back how strongly the drug engages T cells, REGN-5459 seems to be able to lower toxic side effects, but efficacy appears to still be strong. “If you dial down affinity, will that decrease the efficacy? Preclinical models suggest that’s not the case, and we show in this trial responses occurred even at lower doses,” said Attaya Suvannasankha, the abstract presenter and a researcher at the Indiana University Simon Cancer Center. “And at the higher doses, the response rate was 90.5%.”
And the toxicity profile also appeared encouraging. Nearly all patients had only low grade cytokine release syndrome; one patient experienced neurotoxicity. If both the safety profile and efficacy pan out in larger trials, Suvannasankha said, then it’s possible the REGN-5459 could be a more tolerable, effective agent for multiple myeloma.
That could also increase access to myeloma bispecifics, she added. Currently, the need for inpatient treatment for patients starting on approved bispecifics “limits access to patients who may be older, frailer, or far away from treatment centers,” Suvannasankha said.
Genomic profiling finds fewer therapeutic targets in patients with African ancestry
African American patients tend to have the worst burden of colorectal cancer. They have higher incidence and lower rates of survival compared to other populations. One reason for that might be that fewer African American patients have tumors where immunotherapy is more likely to work, according to an abstract presented at AACR.
When researchers sequenced the tumors of over 4,400 colorectal patients at Memorial Sloan Kettering, including 3,265 patients of European ancestry and 245 patients of African ancestry, they found less than 15% of African ancestry patients had tumors with a high tumor mutational burden and microsatellite instable tumors. That was compared to roughly 20% of European ancestry patients. Fewer patients of African ancestry also had actionable alterations in their tumor compared to patients of European ancestry. That means there aren’t as many effective treatment options for many African ancestry patients compared to other ancestry groups.
The difference is largely driven by the fact that the vast majority of genomic research has failed to include large populations of non-European ancestry patients. “The hope is that with more diverse populations in our research studies, we can be powered to find new targets that might benefit these patients,” said Henry Walch, a computational biologist at Memorial Sloan Kettering and the abstract presenter.
HPV vaccine prevents cancers — but the public is forgetting that
HPV causes not only cervical cancer but also head and neck, penile, anal, and vaginal cancer. If you’re reading this, you probably know that. But, as an abstract presented at AACR showed, most Americans still don’t – despite there being a perfectly good vaccine that can prevent HPV since 2006.
Not only that, the fraction of Americans who know that HPV can cause cervical cancer is dropping. “People who know HPV causes cervical cancer has been decreasing since 2017. It decreased by 7.4% to 70.2% of people,” said Eric Adjei Boakye, the presenter of the abstract and a researcher at Henry Ford Health. Boakye looked at data surveying Americans nationally from 2014 to 2020. The number of people who knew HPV caused oral and penile cancers hovered around 30% that whole time.
HPV related cancers are entirely preventable by the HPV vaccine. “We can completely eliminate cervical cancer by getting everyone vaccinated who should be,” said Monica Bertagnolli, the director of the NCI, after the presentation.
Clinicians, Boakye said, should make sure that their patients know that.
Neoadjuvant vs. adjuvant immunotherapy for NSCLC — why not both?
Patients without targetable mutations like EGFR or ALK in non-small cell lung cancer have faced long odds for a long time. Adjuvant chemotherapy, a mainstay of care for these patients, offers only a grim additional 5% survival in five years – but the recent addition of anti-PD1 or PDL1 immunotherapy either before or after surgery have separately offered patients a big jump in survival.
Now, new results from the Phase 3 AEGEAN trial signals that using the anti-PDL1 drug durvalumab both before and after surgery might bring patients yet deeper responses. The trial showed that adding durvalumab before surgery and durvalumab plus chemotherapy after surgery cut the risk of disease recurrence or progression by 32% compared to neoadjuvant chemotherapy plus placebo after about a year.
Pembrolizumab offers modest gains for advanced bile duct cancers
Biliary tract cancers are rare, but tend to be highly aggressive and are often fatal. While few patients respond to immunotherapy drugs alone, new results to the Phase 3 Keynote-966 trial show that combining the immunotherapy drug pembrolizumab with chemotherapy can eke out a modest, but significant, benefit for patients with advanced biliary tract cancer. Patients on the combination saw an average duration of response of 9.7 months, a small benefit over the 6.9-month average duration of response on chemotherapy alone.
Still, any improvement is needed in this setting. Following the results of the TOPAZ-1 study, the FDA approved the addition of durvalumab to chemotherapy for locally advanced or metastatic biliary tract cancer. That study showed an overall survival of 13.5 months on the combination compared to 11.5 months on chemotherapy alone.
STAT will be back with pop-up newsletters at future cancer conferences. Sign up here for ASCO and here for ASH. And sign up here for our twice-weekly Cancer Briefing.