MADRID — Daiichi Sankyo and ADC partner AstraZeneca finally have some clarity on the deaths that occurred in a pair of phase 3 studies for their Enhertu follow up Dato-DXd.
The gist of the safety update, presented in a late-breaking session at the European Society for Medical Oncology (ESMO) Congress in Madrid on Monday afternoon, is that more patients in the lung cancer study experienced interstitial lung disease (ILD) when taking datopotamab deruxtecan (Dato-DXd), compared to breast cancer.
The companies, oncologists at the ESMO meeting and investigators now think that the nature of the damage to the lungs coming into the trial might put a person at greater risk of the adverse event.
Daiichi’s Mark Rutstein, M.D., head of global oncology clinical development, told Fierce Biotech that the benefit-risk profile still seems to tip towards Dato-DXd, particularly in patients with non-squamous non-small cell lung cancer (NSCLC). He called the deaths unfortunate.
“I cannot think of anything more important than the patients’ safety, hands down,” he said. Daiichi has therefore worked to educate physicians on how to reduce the severity of ILD, which causes scarring of the lungs and can lead to irreversible lung damage.
AstraZeneca’s Susan Galbraith, Ph.D., executive vice president of oncology R&D, said in an interview that the rate of adverse events was low in both Dato-DXd studies, which did better on tolerability than chemotherapy. There was also a low discontinuation rate.
Daiichi and partner AstraZeneca presented two Dato-DXd studies during the presidential session at ESMO Monday afternoon. The first, TROPION-Lung01, compared the therapy to the current standard of care, the chemotherapy docetaxel, in patients with locally advanced or metastatic NSCLC who had previously received at least one prior line of therapy.
The second study, called TROPION-Breast01, compared Dato-DXd with investigator’s choice of chemotherapy in patients with inoperable or metastatic HR+, HER2 low or negative breast cancer who had previously received endocrine-based therapy and had at least one systemic therapy.
Picking up on patterns
In TROPION-Lung01, 3% of patients in the Dato-DXd arm experienced ILD compared to 1% in the docetaxel arm. There were seven deaths, or grade 5, ILD events that were determined to be drug-related by an independent committee. In four of the cases, the primary cause of death was considered by the investigator to be disease progression. Of the seven deaths, four patients had non-squamous NSCLC and three had squamous NSCLC. The docetaxel group had one death attributed to ILD.
Meanwhile, in TROPION-Breast01, the rate of ILD was 3% in the Dato-DXd arm and the severity of these events was low. There was, however, one death that was deemed drug-related by the committee, which had a primary cause of disease progression.
“ILD seems to be more prevalent in the lung cancer population. It kind of makes sense, right? Because these patients have an insult to the lungs,” Rutstein said. These patients also typically have a history of smoking. “We’re picking up on these patterns so that we can even further improve and mitigate the severity of ILD.”
ILD is a known side effect and has previously occurred with Enhertu, as well. “Actually, that’s where some of our learnings come from,” Rutstein noted.
Memorial Sloan Kettering breast oncologist Sarat Chandarlapaty, M.D., Ph.D., did not seem overly concerned about the ILD events in the lung cancer trial, especially as the investigators deemed them to be more disease-related than treatment-related. He said the data show a “very low rate of ILD.” In the breast cancer study, Chandarlapaty, who is the Naddisy Foundation Chair in breast cancer research at MSK, noted the low rate of ILD and more typical side effects such as nausea and vomiting.
Another factor may be the level of pre-treatment, although this is still what Rutstein called a hypothesis. If patients had undergone many chemotherapy treatments prior to arriving in the study, they seemed to be at greater risk of ILD.
“We will do everything we can to continue to mitigate that risk but the important aspect is, again, when we step back and look at any of the datasets, it always has to be about balancing the benefit and the risk,” Rutstein said.
Still waiting for OS
As for efficacy, in the lung cancer study Dato-DXd reduced the risk of disease progression or death by 25% compared to docetaxel. The median survival was 4.4 months compared to 3.7 months on chemotherapy. Daiichi also reported a confirmed objective response rate of 26.4% in those who received Dato-DXd, compared to 12.8% of the patients who received chemotherapy.
When looking at the entire population, both squamous and non-squamous disease, Chandarlapaty said, “There was a statistically significant benefit, but it was not numerically a very large difference.” When the results were broken out by disease type, patients with non-squamous disease fared better with a two-month improvement in progression free survival.
Chandarlapaty isn’t sure why there is a difference between these two patient groups.
“But it does sort of provide, I think, a path forward in terms of thinking about which population is most likely to benefit. Looking ahead, I think this would represent a meaningful second-line standard for the non-squamous,” he said.
AstraZeneca’s Galbraith is particularly thrilled with the non-squamous results, which she called “easily identifiable” in the data. She believes Dato-DXd may be able to replace chemotherapy in the future.
One thing the ESMO crowd will not get to see is the hotly anticipated overall survival data for either TROPION program. In July, the partners disclosed that the data were not yet mature. That, plus the disclosure of the patient deaths, sent AstraZeneca’s shares down about 5%.
Rutstein said the situation has not changed, and neither the lung nor breast data is mature on this primary endpoint yet. The OS data have not yet crossed the threshold for statistical significance, he explained.
Daiichi reported that the OS interim results “numerically favored” Dato-DXd over chemotherapy in the breast cancer study, with a hazard ratio of 0.84. So the study will have to proceed to the final analysis before that endpoint is clarified.
“The good news is that so far, although it’s very early, it heads in the right direction,” Rutstein said.
The situation was the same in the lung cancer trial, with results that “numerically favored” Dato-DXd over docetaxel in the overall population with a hazard ratio of 0.90. For patients with non-squamous tumors, the hazard ratio was 0.77. These results did not reach statistical significance as of the cut off. Rutstein said some of the TROPION-Lung01 overall survival data will be presented this afternoon at ESMO.
The fact that OS has not been reached in either trial is “no surprise,” Rutstein added. While he couldn’t speak to specifics, Rutstein did say that this has been discussed with regulators. The hope is that the agencies will consider the larger Dato-DXd data package and consider all the benefits and risks before making a regulatory decision.
“Certainly, overall survival is an important endpoint, however, there are many different things to consider,” he said. “Having done this for many years, when regulatory agencies consider whether or not they’re going to approve a regimen for patients, they’re going to look at the totality of the data. You’re not just going to look at one endpoint, and they’re going to look at that overall benefit-risk assessment to make sure that benefit outweighs risk.”
In situations of high unmet need, sometimes OS is not needed to secure an approval. Rutstein is confident the efficacy they are showing will bring a compelling data package.
For instance, when Enhertu was considered for mutation-positive NSCLC, agencies assessed phase 2 trials in that indication with the data focusing on response rate, durability and “some survival data.” Daiichi and AstraZeneca have yet to produce phase 3 evidence, as the late-stage trial is still ongoing in that indication, but Enhertu nevertheless got approved.
“We have confidence in Dato-DXd,” Rutstein said. “These are positive phase 3 trials. We wouldn’t be discussing with the regulatory agency unless we felt there was a favorable benefit-risk profile.”
Chandarlapaty, however, believes having mature OS data could be key to physicians and patients choosing one therapy over another. There is another approved TROP2 antibody in use now that has mature OS data. The two have differing adverse events but it’s still hard to compare the two without Dato-DXd’s numbers.
“These are differences that aren’t so easy to say why I would choose one over the other. I think at the moment, just knowing that you have mature overall survival data might sway someone if they had a choice,” Chandarlapaty said.
Daiichi and AstraZeneca are keeping the timeline for the final analysis fairly close to the vest. For the lung data, AstraZeneca has disclosed a timeline of “sometime in 2024.” The breast cancer timeline has not been disclosed.