In this study, prospective outcome data from a cohort of 373 patients whose adjuvant therapy decision was based on risk stratification with the multigene test EndoPredict are presented. To our knowledge, these are the first prospective clinical outcome data for the multigene test EndoPredict.
The most important finding is the fivefold increased risk for distant metastases in EPclin high-risk patients versus EPclin low-risk patients (HR 5.18, p = 0,04). This is in line with formerly published data from retrospective analyses using EP as a prognostic biomarker in ER-positive, HER2-negative patient cohorts; EP could be shown to be an independent predictor of distant recurrence based on retrospective analyses of prospectively collected clinical data from the ABCSG6&8 and TransATAC [10, 15]. The hazard ratios for distant recurrence in EPclin high-risk versus EPclin low-risk patients in these studies were reported as 4.77 and 5.99, respectively, and thus lie in the range of our findings.
In our study, EPclin allocated almost two thirds of the patients into the low-risk group while one-third was classified as being at high-risk for distant recurrence. This is consistent with other previously published real-world data: A French prospective multicenter trial on decision impact included 201 patients, of which 67% were classified as low risk by EPclin . A smaller German study of 82 cases reports an EPclin low-risk rate of 68% . This underlines the robustness of the EndoPredict test in clinical routine.
In this study, we also collected data on patients’ compliance with the adjuvant therapy recommendation given during individual case discussion in the interdisciplinary tumour board. After a median follow-up of 3.5 years, 14% of patients in our study were noncompliant regarding the recommended ET. This compliance rate corresponds with data, where non-adherence to adjuvant ET is reported as being very common, with estimates of up to 50% of patients not successfully completing a 5-year course of treatment [18,19,20]. Retrospective studies based on databases and registers have shown adherence to adjuvant endocrine therapy of approximately 60–82% and 46–73% after 3 and 5 years, respectively [21, 22].
Since adjuvant CTX leads to significant outcome improvement in endocrine-sensitive breast cancer , we analysed whether patients with recommendation for adjuvant CTX actually did receive this treatment. Of the 128 patients who were recommended to undergo CTX in addition to ET, 36 patients did not receive adjuvant CTX, accounting for a CTX non-compliance rate of 28%. Even though this rate of chemotherapy non-compliance at first glance might deem to be surprisingly high, it complies with existing evidence from the German prospective multi-center study BRENDA II. Here 28% of the patients who were allocated to an intermediate-risk group according to conventional clinicopathological prognostic factors, and were recommended to undergo adjuvant CTX by the interdisciplinary tumor board, did, in fact, not adhere to this recommendation and thus did not receive adjuvant CTX .
To further elucidate the potential benefit of adjuvant CTX in EPclin high-risk patients of our study cohort, we compared DFS of patients who did, versus did not, receive the recommended adjuvant CTX. We found a remarkable 68% reduction in the risk for death or relapse after 3 years by administration of adjuvant chemotherapy (HR 0.32; 95% CI 0.10–1.05; p = 0.06). These data can be interpreted as a prospective confirmation of a study latterly published by Sestak et al.: within a comparative analysis of 3746 patients, they showed that the rate of increase in risk for distant recurrence in patients with high EPclin score can be reduced by administration of adjuvant CTX in addition to ET . Thus, both our and Sestak et al.’s study give evidence that EP indeed can be used to identify patients who might benefit from chemotherapy in ER-positive, HER2-negative breast cancer. Conversely, EP allows the identification of a group of patients who can safely forgo the use of adjuvant CTX; in our study, patients with EPclin low risk experienced a favourable 3-year DMFS-rate of 99.6% (95% CI 98.7–100%). It seems to be unlikely that longer follow-up will result in a substantial change of this positive outcome without adjuvant chemotherapy since the main efficacy of adjuvant chemotherapy is known to evolve during the first 3–5 years after diagnosis . Recently published data on long-term prediction of distant recurrence using EP support this: Filipits et al. reevaluated the combined ABCSG-6/8 cohorts with longer follow-up and reported a 15-year distant recurrence-free rate of 93.4% for EPclin low-risk patients .
There are certain limitations to this study. They include the relatively short follow-up period, the moderate sample size and the monocentric, non-randomized design. The discussed early results need to be confirmed with longer follow-up, which will be available in the near future. The results then might also support existing evidence from published retrospective studies showing that EP as a second-generation multigene test is able to extend its prognostic value to prediction of late recurrences [7, 15]. Of note, the international, multicenter study “RESCUE” (“Reaching for evidence based chemotherapy use in endocrine-sensitive breast cancer”, NCT 03503799) has started recruitment in Germany . The study documents distant metastasis-free survival, disease-free survival and overall survival events in patients whose tumors had been tested with EP, and will allow further confirmation of our results.