Do tumor infiltrating lymphocytes (TILs) predict benefits from trastuzumab therapy for HER2 positive breast cancer? Meta-analysis of individual patient data from 4097 women in 5 trials.
Hills RK., Bradley R., Braybrooke J., Gray RG., Taylor H., Denkert C., Badve SS., Kim RS., Lacroix-Triki M., Regan MM., Hayes DF., Dowsett M., Tutt ANJ., Gelber RD., Cameron DA., Bergh JCS., Swain SM., Michiels S., Loi S., Salgado R.
508 Background: High TIL counts are associated with a lower risk of breast cancer recurrence, especially in women with ER negative, HER2 negative tumors and, possibly, greater benefit from trastuzumab in women with HER2 positive cancer: the FinHER trial reported a differential effect of trastuzumab based upon TIL status. We performed a meta-analysis of randomized trials of trastuzumab in early breast cancer to attempt to validate this finding. Methods: TILs were quantified in 4097 women in 5 randomized controlled trials (NSABP B-31, FinHER, HERA, Intergroup N9831, PACS-04). All trials contributed to the Early Breast Cancer Trialists' Collaborative Group individual patient data meta-analysis of trastuzumab for women with HER2 positive tumors which found a significant benefit for trastuzumab therapy. TILs were assessed using established International Guidelines, with HERA using digital TIL-scores. The primary outcome was time to first recurrence. Cox regression analyses, adjusted for trial, treatment allocation, and nodal status, were used to quantify the prognostic value of TILs; and standard stratified logrank tests were used to assess the differential effect of trastuzumab therapy. Results: The median percentage TILs was 13% (Interquartile Range 5-30), with fewer than 10% of patients exhibiting TILs >50%. The prognostic value of TILs was confirmed, with patients with higher TILs being at lower risk for recurrence (adjusted hazard ratio per 10% increase in TILs 0.87 (95% CI 0.84-0.90), p<.0001) with similar effects in both treatment groups. Outcomes improved steadily with increasing TILs, and unadjusted 10-year recurrence rates fell from 30% in women with TILs <10% to 15% in those with TILs 70% or greater. Consequently, analyses of the predictive effect of TILs were stratified into 5 groups (0-9, 10-19, 20-39, 40-59, 60+). Overall, there was a highly significant benefit of trastuzumab on recurrence (HR 0.62 (0.54-0.70) p<.0001), but there was no evidence of any interaction between TILs and the proportional reduction in recurrence (p=0.8 for heterogeneity and trend). Conclusions: While higher TILs are associated with lower recurrence rates, there was no indication that the proportional reduction in recurrence with trastuzumab varied by TILs, although the number of patients with high levels was limited. Owing to a lower underlying recurrence rate, absolute benefits from trastuzumab were lower, but still substantial, in women with high TIL tumors. Clinical trial information: NCT00045032 , ISRCTN76560285 , NCT00005970 , NCT00004067 , NCT00054587 . [Table: see text]