Dr. Joseph Sparano, 2024 OncLive Giant of Oncology in Breast Cancer, presents update on results from the TAILORx breast cancer trial at San Antonio Breast Cancer Symposium 2022
And it's my pleasure to introduce our next speaker. That's Doctor Joseph Spiro from the Icann School of Medicine at Mount Sinai. And he will be presenting the trial assigning individualized options for treatment or tailor X. An update including 12 year event rates. Doctors, Doctor Goel, doctor Partridge um on behalf of the Taylor X investigators. It's my privilege to present an update of the Taylor X trial including a 12 year event rates. This NC I sponsored trial was coordinate by the E Cog Aquin Cancer Research Group in collaboration with all other Cooper groups and was supported with funding not only from NC I but also from the Breast cancer Research Foundation, Coleman Foundation and the US Postal Service Stamp Fund. The study design and accrual are depicted here after eligibility verification and consent study volunteers were preregistered and had a tumor specimen sent to the genomic health laboratory for the recurrence score assay upon receipt of the results by the local site, subjects were then registered and treatment assignment or randomization uh was uh directed as shown 10,002 hun 273 volunteers were registered between April of 2006 and October of 2000 10 resulting in 6711 valuable subjects or 65% in the recurrent score. 11 to 25 group uh randomized to receive endocrine therapy plus chemotherapy. Um The standard arm or endocrine therapy alone, the experimental arm, the remaining 35% were assigned to receive endocrine therapy alone. If the recurrent score was 0 to 10 and chemo plus endocrine therapy. If the recurrent score was 26 to 100 the trial employed a noninferiority design for the recurrent score 11 to 25 arms with invasive disease free survival as the primary endpoint and with a margin, a hazard ratio margin of 1.332. Um as a uh conclusion of non inferiority, the analysis we were conducted at full information after 836 disease free um events um in the recurrent score 11 to 25 group after a meeting follow up of 7.5 years. The main study findings included the following for the recurrent score, 0 to 10 group assigned to endocrine therapy alone. Distant relapse free interval rates were 99% in five years and 97% in nine years. For those with the midrange recurrent score of 11 to 25 endocrine therapy was non inferior to chemo endocrine therapy for invasive disease free survival and distant re relapse free interval. The primary and secondary end points respectively, relapse free interval. And overall survival were also similar for those with a high recurrent score of 26 to 100 assigned to endocrine therapy. High recurrence score was prognostic for high recurrence risk. The estimated absolute chemotherapy benefit ranged from about 8% if the recurrent score was eight to excuse me, 3026 to 30 approximately 30% if the recurrence score was greater than 30 based on the estimate uh estimated chemotherapy benefit observed in the B 20 prospective validation trial. Other key study findings that were secondary or exploratory included the following. First, clinical pathologic factors such as tumor size, uh grade and age and provide added prognostic information to recurrent score but not predictive information for chemotherapy benefit. Second, there was some chemotherapy benefit noted for women with a recur up to 50 years of age with a recurring score of 16 to 25. Third, integration of the recurrent score plus the clinical pathologic features yielded more prognostic information than either alone leading to the development of the online Rs Clin tool. And finally, additional post hoc exploratory analysis revealed that although black race was associated with worse outcomes in other races, recurrence score was still prognostic for recurrence and predicted for lack of chemotherapy benefit differences in early discontinuation of endocrine therapy. Clinical pathologic features, insurance coverage and the neighborhood de uh deprivation index did not fully explain these disparities. The rationale for conducting the updated analysis as shown in this slide first and foremost, the annual recurrence risk for er positive her two negative disease persists for up to 20 years or longer. In fact, at least one half of recurrences develop more than five years after diagnosis which has been termed late recurrence. The Taylor X design pres specified follow up for up to 20 years in order to assure adequate follow up for late relapse. The median follow up at the time of the original Taylor X report was 7.5 years and the median duration of endocrine therapy was 5.1 years. Updated analysis at this time provides substantially more information. Regarding late recurrence risk, there was a longer median follow up out to 11 years in the randomized group of with a recurrence score of 11 to 25 and 10.4 years in the overall population. With this additional follow up, there are also substantially more events including all arms compared with the primary analysis including 50% more distant recurrence events in the recurrence score 11 to 25 group. The methods for this analysis are shown here. The same I TT population previously reported in the primary analysis was also used. Event free rates were estimated using the Captain Meyer method. And results displayed uh including the standard error hazard ratios were estimated using partial likelihood uh analysis of the cox proportional hazard mo model self reported race and clinical risk were also examined as covariates. There were no corrections for multiple comparisons. Captain Meyer curves for the recurrent score 11 to 25. Arms are shown here as in the primary analysis. There were no significant differences in the updated analysis when comparing endocrine therapy and chemo endocrine therapy for invasive disease free survival, distant relapse free interval, relapse free interval or overall survival. Five and 12 year event rates for the recurrent score. 11 to 25. Arms are shown here. Distant recurrence rates were about 7% at 12 years. Irrespective of chemotherapy use, averaging a rate of about 0.5% per year late exceeded early recurrences, relapse free interval rates at five and 12 years were 97 and 90% respectively. With a similar pattern for distant recurrence rates and non recurrence events reflected by the difference between the invasive disease free survival, endpoint and relapse free interval rates occurred in 13% of 12 years. Averaging about 1% per year. Overall survival rates were similar with most, most deaths occurring after five years. The Captain Meyer curves for all four end points and all trial arms are shown here for the overall population recurrent score was prognostic for all four end points with a recurrent score of greater than 25 displaying the most discriminatory um evidence for prognosis as shown by the Captain Meyer curves for the recurrence score of 0 to 10 group treated with endocrine therapy alone. Distant and overall recurrence free interval rates at 12 years were 93 and 91% for the recurrent score, 11 to 25 group. There was a less than 1% difference for all endpoints. When comparing endocrine therapy with chemo endocrine therapy at 12 years for the recurrent score of 26 to 100 group event rates were substantially higher. Uh when compared with the lower recurrent score, despite the use of adjuvant chemotherapy. In addition to endocrine therapy shown here are the DRF I and RF I event rates for the recurrent score, 11 to 25 arms and women 50 or younger. With the comparisons of endocrine therapy with chemo endocrine therapy stratified by recurrent score within this range. There was no chemo benefit in the recurrent score 11 to 15 group. As shown here with uh dr five rates um exceeding 95% at 12 years. With endocrine therapy alone, there was a marginal chemo benefit in the recurrent score, 16 to 20 group with a less than 1% difference in DRF I at 12 years. There was uh there was benefit uh evident for the recurrent score 11 to 25 group with the absolute difference of about 8% for DRF I at 12 years. The absolute benefit associated with chemotherapy is a function of the relative uh reduction associated with chemotherapy use and the underlying risk of recurrence. In addition, for premenopausal woman chemotherapy induced menopause drives some chemo benefit. In order to disentangle these factors driving chemo benefit, we further evaluated the impact of the recurrent score and clinical risk on the absolute differences of 12 years for distant recurrence rates associated with chemotherapy use in women, 50 or younger with a recurrent score of 16 to 25. When stratified by age and menopausal status in the forest plots. There was a statistically significant three way interaction between chemo use age and recurrent score for invasive disease free survival. Although the interaction P value was not significant for distant recurrence. The forest plot uh showed chemo benefit and preventing distant recurrence for women. 46 to 50 premenopausal in the age at which chemotherapy associated menopause is most likely to occur. We use the binary clinical risk categorization employed in the mind. A trial where low risk was calibrated to at least a 92% tenure, breast cancers specific survival for endocrine therapy alone. Based on version 8.0 of adjuvant online, low risk was defined as a tumor up to one centimeter and high grade two centimeters in intermediate grade or three centimeters in low grade and high risk was defined as not meeting the low-risk criteria. When stratified by clinical risk. There was an absolute 7.8% uh chemo benefit in the recurrent score 21 to 25. In contrast, there was only a 0.4% absolute chemotherapy benefit for a current score of 16 to 20. When further stratified by clinical risk, there was no benefit for low clinical risk patients and a recurrent score of 16 to 20 and 3.1% for the same recurrent score group and high clinical risk. But for the recurrence score of 21 to 25 group, there was a 5.9% of benefit for low clinical risk and 11.7% benefit for high clinical risk. Shown here the outcomes by time period after diagnosis and also by self reported race, The estimated annual hazard rate for distant recurrence was re relatively stable over 12 years, averaging about 0.5% annually for the other end point end points, annual rates were higher after five years including um overall uh relapse including local regional and distant relapse um and death when we evaluated outcomes by race and and time period and models adjusted for other factors. Black race was associated with about a twofold and statistically significantly higher event rate for all events in years, 1 to 5 but not years six through five in models unadjusted for other factors. The adverse impact of race largely held up in this early period but was modestly attenuated when the models were adjusted for age tumor size, grade recurrence score and insurance status. In conclusion, in this updated uh analysis of Taylor X, there was longer median follow up and more events in the randomized group. The main study findings remain unchanged for the recurrent score. 11 to 25. Uh arms endocrine therapy was non inferior to chemo endocrine therapy for invasive disease free survival. The primary end point and distant relapse free interval, the secondary end point relapse free interval and overall survival rates were also similar. Other exploratory key study findings were also similar to the original analysis. There was chemo benefit for women 50 or under with a recurrent score of 21 to 25 that was substantial. There was also some chemo benefit for women 50 or under with a recurrence, recurrent score of 16 to 20 high clinical risk. New findings of this updated analysis which should be regarded as exploratory include the fact that uh late recurrence beyond five years uh exceeded early recurrence. Although that's highly consistent with previous work and um racial disparities for black women were associated with early but not late recurrence. I once again would like to um acknowledge and thank uh the funders including the NC I, the US Breast Cancer Research Stamp Fund, Breast cancer Research Foundation, the Susan G Coleman Foundation, uh the advocate community that supported this trial. The multiple collaborating groups named here, the participating sites including um physicians, nurses, physicians, assistants, research coordinators and assistants and our collaborators at genomic health. And most importantly, I'd like to thank the 10,273 volunteers, Taylor X volunteers. Um Some of whom I treated had the privilege of treating are shown here. Thank you for your attention. Thank you very much, Doctor Spiro for that excellent presentation. And we look forward now to questions from the audience. We'll start with microphone four, please. Uh Hh Bernstein from Boston Joe. Fantastic to see the follow up. And as it relates to the previous presentation on cognitive dysfunction, uh the one way to prevent chemo induced cognitive dysfunction is not to give unnecessary chemotherapy. So it's phenomenal that we can avoid that for premenopausal women, you've consistently shown that women less than age 40 don't seem to benefit from chemo whereas women in their forties might. And the only way I can square that is to imagine that the vast majority of the difference is from chemotherapy induced menopause. Is there any way to tease it out? Uh I know you did the three way thing but is there any way to tease it out more concretely than that at this time? With the, with the information that we have? I don't think so. Um I think we're truly in a situation of um equipoise and I think it provides really sound justification for the Planned NRG trial that will randomize patients who fall in these equipoise groups to chemo endocrine therapy versus endocrine therapy alone. I I agree with your assessment. I do have concerns that those women who are 40 or under who um participated in this trial may have been highly selected. And so for that reason, I'm, I'm, I'll be a little concerned about drawing um that conclusion that those patients truly don't benefit. Uh Microphone five please, Eric Weiner, uh Joe, if there's no benefit overall and yet there's benefit in the premenopausal subset. Does that mean there's harm in the post menopausal women associated with chemotherapy? Um No, there does not appear to be based on the uh stratified analysis looking specifically at, at the post menopausal group. Um There was only, you may recall that there was really no difference in the overall population. Um About two thirds of the patients in the trial were post menopausal about or or older than 50 about um one third were 50 or under. Um And the group with a recurrence score of 16 to 25 represents only a fraction of the younger group. So um I think when you, when you um and, and for that reason, it, it just, it, it, it has a really overall small effect on the entire uh randomized comparison. Thanks. Thank you microphone three University College London and interesting data. The EBC TCG meta analyses of 2012 and 2019 show that in patients with positive her two negative, sorry, positive breast cancer, there are very few recurrence events after five years that can be influenced by chemotherapy and almost none after about 7.5 years. And do you think that's data affects your conclusions in any way? I think this data uh is reassuring evidence that we would, we wouldn't see um that we uh see a lack of, of of benefit or or beyond five years. Um Our data are really consistent with that. Um However, since we, we, we didn't really see a benefit for chemotherapy in the overall population. Um with shorter follow up, we, we didn't really expect to see benefit emerge with longer follow up. But it, it's still reassuring to see that with this long degree of follow up, the the primary trial conclusions remain unchanged. We have time for one more question. A microphone. Four, please. Thank you, Joe. Uh Matthew gets from Mayo Clinic. Just excellent presentation question is we, we have the, we have the date of Mark's ponder where we see benefit of chemotherapy regardless of the recurrent score in, you know, the premenopausal patients are aged less than 50. Here, we're, we're looking at, you know, slight differences in the recurrent score and we're saying that one group gets it versus not, what's the chance that this is just chance? And really at the end of the day, the proportional reductions are the same, but we're, you're just enrolling such a good risk population. So we're, we're unable to see the differences. I agree with, I agree with your point. Um It's, it's difficult to disentangle the the um effects of chemotherapy that are are, are driven the beneficial effects that may be driven by bringing on early menopause versus a true cytotoxic effect in eradicating micro metastatic disease. Um and the, the, the, the discordance that we see in the two trials can be exactly what you're suggesting that it may be just a function of the higher clinical risk and the Rx Boer patients that do allow you to, to detect a marginal uh benefit from chemotherapy that's being driven by uh early induction of menopause. Thank you very much, Doctor Espero for your presentation. Once again, it's clear from the number of questions we also received online um that there's a lot of interest in that presentation.
