Abstract P1-06-06: No age-related outcome disparities according to 21-gene recurrence score groups in early breast cancer patients treated by adjuvant chemotherapy in the prospective WSG PlanB trial

Background: Elderly breast cancer (BC) patients (pts) have been reported to have worse BC-related outcome than younger pts, even within clinical trials such as TEAM. Shak et al. recently showed in a large SEER data analysis that in the high 21-gene recurrence score (RS) group, older pts (>70y) receive less chemotherapy (CT) and have a worse BC-specific mortality than younger pts. Here, we therefore aimed to see whether there are age-related outcome disparities according to RS groups in pts receiving state-of-the-art CT in the prospective WSG PlanB trial.

Material and Methods: PlanB compared 6 cycles of anthracycline-free TC vs. standard anthracycline-taxane based CT (4xECà4xDoc) in patients with high risk pN0 (T2-4, G2-3, <35 years, or high uPA/PAI-1) or pN+ HER2- early BC. 21-gene assay was performed in all HR+ tumors and omission of chemotherapy (CT) recommended in RS≤11 HR+ pN0-1 BC. Final analysis for the CT randomization for RS 12-25 after 60 months median follow-up revealed similar 5-year DFS and OS outcomes for both CT arms (ASCO 2017).

Results: In all pts with luminal cancer and RS results (n=2577), there was an age-related significant difference in RS risk group assignment (p<0.0001): in young pts (<40y), 9.1% had RS≤11, 52.7% RS 12-25, and 38.2% RS>25; in pts 40-69 years, 18.3% had RS≤11, 61% RS 12-25, and 20.7%% RS>25; in elderly pts (>70y), 19.5% had RS≤11, 55.3% RS 12-25, and 25.2% RS>25. Among patients receiving chemotherapy, RS>25 vs. RS<25 was associated with significantly poorer DFS separately within the elderly subgroup (HR=3.03, 95%-CI [1.15-7.96]) and in those aged 40-69 years (HR=3.14, 95%-CI [2.18-4.52]); there were only nine events among patients <40y. In particular, among pts receiving CT with RS>25, there were no significant differences in DFS between any two of these three age groups.

Conclusion: A substantial percentage of elderly patients (> 70y) presents with high-risk luminal disease; these patients are candidates for CT. In PlanB, about 25% of elderly luminal BC patients had high-risk (RS>25) tumors. Nevertheless, after receiving modern adjuvant CT, their DFS was comparable to that of non-elderly pts with high-risk RS tumors. Consequently, older BC pts with high-risk luminal tumors who are fit enough to receive adjuvant CT should be treated according to guidelines in order to overcome age-dependent survival disparities which have been observed in registries for high-RS tumors.

Citation Format: Harbeck N, Gluz O, Wuerstlein R, Clemens M, Malter W, Reimer T, Nuding B, Stefek A, Pollmanns A, Augustin D, Uleer C, Lorenz-Salehi F, Shak S, Chao C, Christgen M, Kates R, Kreipe H, Nitz U. No age-related outcome disparities according to 21-gene recurrence score groups in early breast cancer patients treated by adjuvant chemotherapy in the prospective WSG PlanB trial [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P1-06-06.

Abstract P2-07-01: Association of TILs with clinical parameters, recurrence score, and prognosis in patients with early HER2-negative breast cancer (BC) – A translational analysis of the prospective WSG planB trial

Introduction:

Tumor-infiltrating lymphocytes (TILs) have been associated with prognosis and with chemotherapy response among patients with BC, particularly in presence of high-risk features. The WSG planB trial randomized 2448 patients with HER2- N0/1 BC for comparison of anthracycline-free (6xTC) vs. standard anthracycline-taxane chemotherapy (4xEC-4xDoc). Recurrence Score® (RS) was incorporated for risk stratification in hormone receptor positive (HR+) BC. The present analysis focuses on the correlation of TILs with clinical/pathological parameters and their prognostic impact among planB patients.

Methods:

Stromal TILs were evaluated using a pathologist and two-observer approach. Three independent observers evaluated digital sections on H&E staining as previously suggested (Salgado et al., Ann Oncol. 2014); the median of the three values (TILmed) was used for statistical analysis. Spearman correlations of TILmed with clinical/pathological parameters (including central KI67 expression, quantitative ER measurements, nodal involvement, and RS) and univariate impact on event-free survival (EFS) were analyzed.

Results:

Our analysis included 300 patients with HR- and 1124 patients with HR+ HER2- BC. Both in HR- and HR+ BC, a significant association between TILmed and (i) central grading (correlation coefficient r=0.147, p=0.012 and r=0.195, p<0.001, respectively) and (ii) central Ki67 expression (r=0.202, p=0.001 and r=0.152 and p<0.001) was observed. Among HR+ cases, a significant association between TILmed and quantitative ER measurements (r=-0.412, p=0.041) and RS (r=0.190, p<0.001) was found. Furthermore, univariate Cox analysis revealed a significant association between TILmed (coded as fractional rank) and event-free survival (EFS). The hazard ratio of 75th to 25th percentile was 1.58 (95%CI: 1.06-2.36, p=0.025). This impact was not separately significant in HR subgroups due to lack of events

Conclusion:

In this dataset, presence of stromal TILs was moderately associated with clinical features of high-risk breast cancer (including RS) and decreased EFS. TILs will be evaluated as a prognostic or predictive factor (in multivariate and subgroup analyses) when the outcome results are evaluated after prolonged follow up. Furthermore, an updated analysis including the complete planB dataset will be presented.

Citation Format: Liedtke C, Gluz O, Heinisch F, Feuerhake F, Kreipe HH, Clemens M, Nuding B, Kraemer S, Reimer T, Svedman C, Shak S, Nitz U, Kates RE, Harbeck N, Christgen M. Association of TILs with clinical parameters, recurrence score, and prognosis in patients with early HER2-negative breast cancer (BC) – A translational analysis of the prospective WSG planB trial. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P2-07-01.

Abstract P6-05-11: Run-in phase of prospective WSG-ADAPT HR+/HER2- trial demonstrates feasibility of early endocrine sensitivity prediction by recurrence score and conventional parameters in clinical routine

Background: Despite promising evidence regarding outcome prediction, endocrine sensitivity, as determined by proliferation response to short-term preoperative endocrine therapy, is currently not included in adjuvant chemotherapy decisions in early HR+/HER2- breast cancer (BC).

Methods: The prospective WSG-ADAPT HR+/HER2- trial includes early BC patients with 0-3 positive LN who are candidates for adjuvant chemotherapy based on clinical-pathological criteria alone; it aims to spare chemotherapy in a substantial proportion utilizing a combination of genomic assessment by Oncotype DX and endocrine sensitivity testing. All patients received 3-week preoperative endocrine induction therapy (ET): aromatase inhibitors (AI) if postmenopausal, tamoxifen if premenopausal. Patients with low (0-11) Recurrence Score (RS) or intermediate RS (12-25) and ET response (centrally tested, post-therapy Ki-67 <10%) are recommended to forego adjuvant chemotherapy (“low-risk” patients). Distribution of RS, responder percentages in each group, and impacts of RS, ET regimen, and initial Ki-67 on post-therapy Ki-67 are reported here.

Results: As of 6/2013, 380 patients from 30 study centers had been enrolled in the ADAPT HR+/HER2- trial. Median age was 54 years. At first pre-planned analysis (5/2013), paired Ki-67 measurements (pre-/post-therapy) were available in 241 patients; RS was available in 208 cases (201 with paired Ki-67). RS was low in 21.6%, intermediate in 57.7%, and high in 20.7%; the respective risk group responder percentages (post-treatment Ki 67 <10%) were 84.1%, 73.9%, and 40.0% (p<0.001 when comparing low/intermediate vs. high, chi-square). In particular, these percentages support the pre-trial estimate of >70% endocrine responders in the intermediate genomic risk group, who could potentially be spared adjuvant chemotherapy. Median Ki 67 level decreases (as percentage of pre-treatment value) were 25% in premenopausal patients (tamoxifen, n = 101) vs. 75% in postmenopausal patients (AI, n = 115) (p<0.001, Mann-Whitney); median decreases by RS group were similar, 61% (low), 53% (intermediate) and 56% (high), respectively (p = 0.81, Kruskal-Wallis). In linear regression, pre-treatment Ki-67, endocrine regimen/menopausal status, and RS were all independent predictors for post-treatment Ki 67. Final run-in-phase analysis and validation will be presented after completion of endocrine induction therapy in 400 patients.

Conclusions: The Run-In Phase of the WSG ADAPT HR+/HER2- trial confirms trial design estimates of RS and proliferation response to induction ET. It indicates that the multicenter prospective ADAPT concept combining static and dynamic biomarker assessment for individualized therapy decisions in early BC is feasible. Proliferation response was strongly associated with therapy group (AI/post-menopausal vs. tamoxifen/pre-menopausal). Survival non-inferiority of intermediate Recurrence Score proliferation responders vs. low Recurrence Score patients (active control) will be tested in the ADAPT main phase to determine if adjuvant chemotherapy can be spared in 70% of patients with 0-3 positive LN classified as “intermediate risk” by conventional factors.

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P6-05-11.