Good morning. It’s my pleasure this morning to introduce to you the KATHERINE trial. This is an academic pharma collaboration between NSABP, the German Breast Group and Roche Genentech and a lot of global KATHERINE investigators. KATHERINE is a phase III study of TDM1 versus trastuzumab as adjuvant therapy in patients with HER2 positive early breast cancer who have residual invasive disease following standard neoadjuvant therapy. My disclosures are shown here before I begin.
The rationale for KATHERINE was that we know that outcomes for patients with HER2 positive early breast cancer have been greatly improved with the addition of HER2 targeted monoclonal antibody therapy to chemotherapy. Patients receiving these therapies prior to surgery with clearance of the invasive cancer in the breast and with negative axillary lymph nodes enjoy a favourable prognosis. However, those patients with residual invasive disease in the breast specimen or axillary nodes have a less favourable prognosis with an increased risk of recurrence and death. Since TDM1 is active and approved for patients with HER2 positive metastatic breast cancer following prior therapy with taxane chemotherapy and HER2 targeted therapy which form the basis of our current neoadjuvant therapy, KATHERINE investigated whether substitution of adjuvant TDM1 for adjuvant trastuzumab, which has been a standard, would improve outcomes for these patients with this relative unmet medical need.
The study design was straightforward, this was a global study so we allowed an array of neoadjuvant therapies to be administered to patients before entering the study. We required a couple of things – they had to have central confirmation that they indeed had HER2 positive breast cancer and we wanted to be sure that they had received a standard chemotherapy treatment that we characterised as saying at least six cycles of chemotherapy, there had to be at least nine weeks of a taxane included in there along with trastuzumab to describe, basically, the standard regimens. It was important that all chemotherapy had to be completed before surgery because we were using that biomarker of residual disease as our indicator of relative resistant micrometastatic disease. Of course, we had to have evidence that the patient had residual invasive tumour in the breast specimen or in their axillary nodes and patients entered and began on KATHERINE within twelve weeks of surgery. This is a group of patients with a lot of variability so an important aspect of the study was establishing the stratification factors because patient outcomes and prognosis is different depending on these factors. So we stratified our patients by whether they had originally presented with operable or inoperable disease, by whether they had hormone receptor positive or negative disease, by whether they had single or dual HER2 targeted neoadjuvant therapy and then did they have disease still in their axillary lymph nodes.
Our study randomised 1,486 women one to one to receive either TDM1 or standard trastuzumab every three weeks for fourteen cycles. The primary endpoint for the study was invasive disease free survival. We assumed that adjuvant trastuzumab would provide a three year IDFS of about 70% and then improving that with TDM1 to 76.5% for a hazard of 0.75 would be clinically meaningful. In order to see that difference we thought we would need to have enough accrual so we would have 384 events to give us adequate statistical power and that’s where the sample size of nearly 1,500 patients came from. We did specify that when we had about two thirds of the 384 events we would conduct a single interim safety and interim efficacy analysis which occurred in July of this year and if we saw in that early analysis that our hazard ratio was under 0.73 we would report and also conduct an initial survival analysis.
This is the result of our interim analysis that showed really a remarkable reduction in the risk for developing one of the IDFS events. We had a total of 265 events and you can see 165 of those occurred in the patients receiving trastuzumab, or 22% of that group, and it was reduced to 91 or 12% of patients receiving TDM1. This corresponded to an unstratified hazard ratio of 0.50, substantially lower than what our target hazard ratio had been, which was highly statistically significant. In an absolute sense regarding improvement, three year invasive disease free survival went up from 77% to 88%, a full 11% percentage point increase.
Once we knew that there was this difference it was important to look at how the drug performed across the various subgroups, our stratification, to see if there was consistency. One of the remarkable findings is there really is a striking homogeneity of consistency in terms of the efficacy in all these various subgroups. We see that patients derive the same benefit whether they came in with inoperable versus operable disease, hormone receptor positive or negative, whether they got single or dual HER2 targeted therapy and so on. We saw no differences in efficacy by age and race as well and there are some other subgroups that we will be showing in a presentation. Again, this is a consistent theme, a striking consistency of results that we saw.
In terms of safety we basically knew going in that trastuzumab would be expected to have an increased number of adverse events relative to trastuzumab. Trastuzumab is a well-tolerated drug and we saw this in our study. The important thing is to see that the majority of these adverse events were actually grade 1, grade 2, milder symptoms and these are shown in the lighter colours on these bar graphs. The more severe grade 3 toxicities are coded for in the darker colours and you can see that they’re relatively infrequent. The most common side effects that patients reported on TDM1 were fatigue and nausea so these are certainly things that can be bothersome but are generally manageable and reversible. The potentially serious side effects are platelet count drops and increases in liver enzyme tests shown here. We did see 6% of patients develop more severe thrombocytopenia but more severe liver elevations were really infrequent in the study. So overall this does appear to be a tolerable and manageable regimen.
One other thing I did want to call out was that sensory neuropathy which can be a problem for patients who received taxane-based chemotherapies and TDM1 is we did see an increase in overall sensory neuropathy from 7% to 19% but importantly 12% of that 19%, so to speak, were due to milder grade 1 toxicity and as we’re following these patients we are seeing substantial reversibility.
So in summary, KATHERINE demonstrated that adjuvant TDM1 has both a statistically significant and clinically meaningful improvement in IDFS compared with trastuzumab. We had hoped to demonstrate a hazard ratio below 0.75 and it was actually 0.5, this of course corresponds to a very large and substantial absolute improvement. The benefit of TDM1 for our primary endpoint was consistent across all key subgroups. Safety data were consistent with the known toxicities of TDM1 with expected increases in manageable AEs but these are things that we think would be manageable by a medical oncologist and temporary for the most part.
Additional follow-up will be necessary to evaluate the effect of TDM1 on overall survival but the strength of the data that we’re seeing today makes it likely that the KATHERINE results will form the foundation of a new standard of care in this population and will also substantially increase the use of neoadjuvant therapy in HER2 positive early breast cancer.