At ESMO I presented the Temab-A plus bevacizumab data, which also had a control arm of TAS-102 plus bevacizumab. So Temab-A is an antibody drug-conjugate that targets c-Met with a topoisomerase 1 payload [??]. We’ve previously known that c-Met is up-regulated in a variety of solid tumours associated with poor prognosis, especially in colorectal cancer. We’ve also previously seen monotherapy data from the first in human study of Temab-A, presented at ASCO 2024, specifically for colorectal cancer and it demonstrated promising safety and efficacy signals as a monotherapy. What I presented was another expansion cohort from that first in human study which evaluated Temab-A, so the antibody drug conjugate, plus bevacizumab, which we first went for with a dose escalation and a randomised expansion of Temab-A at two different doses for dose optimisation – 2.4 mg/kg plus bevacizumab, 2 mg/kg plus bevacizumab in that comparative third arm of TAS-102 plus bevacizumab, which would be the standard of care.
What were the results of this study?
So again to highlight, we enrolled patients that were third-line plus colorectal cancer and these patients were unselected for c-Met expression. They had received the appropriate priors of 5-FU, oxaliplatin, irinotecan, appropriate biological therapy, and they had not received prior regorafenib, they had not received prior TAS-102, because that again was the control compared arm. And we allowed enrolment with haemoglobin cutoff down to 8.0, so again this was a really a real-world patient population that was involved in this study, so I think that that’s important.
Some of the early [??] showed we reviewed the safety and toxicity amongst the arms and, by the way, the arms were relatively well-balanced and most patients had received prior bevacizumab in both arms of the study.
When we reviewed the safety data we did see a difference between 2.4 mg/kg and 2.0 mg/kg in terms of dose reductions, dose interruptions with Temab-A, so slightly higher grade 3 adverse events as well. When looking at the collated Temab-A doses and arms, the toxicity profile in terms of reductions, interruptions and discontinuations and so on was not markedly different than what we saw in the comparitor arm of TAS-102 plus bevacizumab. Then when we looked at the toxicity profile specifically we saw what we would expect from the monotherapy data, primarily haematological and gastrointestinal adverse effects in the Temab-A arms, the haematologic adverse events being dose dependent: higher doses, more haematologic toxicity, and the gastrointestinal ones being less likely to be dose dependent.
When looking at all the AEs, all the adverse events, collated together, again not markedly different between TAS-102 and the Temab arms. Because this is an antibody-drug conjugate, it’s also important to highlight the ILD risk of the agent which turned out to be quite low, 5% overall of any grade, and only one patient in the entire study, in this part of the study, with grade 3 ILD and no grade 4 or 5 ILD events.
Efficacy-wise we saw quite high response rates with the Temab-A arms. The 2 mg/kg dose, we saw a response rate of 19% and the 2.4 mg/kg dose of Temab-A we saw a response rate of 30%. In the comparator arm we saw a response rate of 0% with TAS-102 plus bevacizumab. This all resulted in substantially higher disease control rates than what we report with the comparator arm.
Then we, of course, also looked at other measures of efficacy. We looked at progression-free survival, and we saw a numerically higher progression-free survival in both investigational arms, especially in the 2.4 mg/kg arm where the median progression-free survival was 6.8 months in this third-line plus colorectal cancer population. The comparator arm, on the other hand, median progression-free survival, was 4.2 months, and we did not reach the overall survival rate for the 2.4 mg/kg Temab-A arm. The 2 mg/kg Temab-A arm was right in the middle in terms of progression free survival and you can visually look at the curves from the presentation and see the differences amongst 2.4mg/kg, 2 mg/kg and the comparator arms.
Also I think it’s really important to note that we again did not select these patients for c-Met expression, but did look retrospectively at how c-Met expression might have been associated with response and other measures of efficacy. Frankly, all confidence intervals overlapped at the c-Met levels we looked at. That included 25% c-Met expression 2+ or higher, and 10% 3+ or higher. Thus we were unable to draw any conclusions that c-Met expression was required for further investment of the combination of Temab-A plus bevacizumab as it didn’t seem to be a strong predictor of improved efficacy. This is in contrast to the monotherapy data where we did see much more of a signal that c-Met expression was predictive of higher responses and so on, in progression-free survival with Temab-A.
So in conclusion, the 2.4 mg/kg Temab-A dose really demonstrated the best risk/benefit ratio, in terms of balancing toxicities with strong efficacy signals, supporting further development of this treatment in third-line plus colorectal cancer, and arguably at the 2.4 mg/kg dose in an unselected patient population for c-Met.
What do you think is the significance of these results, and what is next for this study?
The significance of these results is we’re demonstrating a similar adverse event profile to the typical standard of care arms we do, and numerically substantially higher response rates in median progression-free survival. So this has the potential, certainly, to be evaluated further in a much larger and potentially registrational study, where these would be compared directly head-to-head. What is ongoing right now is a phase III clinical trial evaluating the monotherapy Temab-A for 10% or greater 3+ c-Met expression, and that’s the first large-scale clinical trial we’ll see with this drug in the third-line plus colorectal cancer patient population. But, again, really the data that I presented supports further development of adding bevacizumab onto that in an unselected patient population.
