Edith Perez, MD
Among HER2-positive breast cancer patients treated with chemotherapy alone, women with high levels of stromal tumor-infiltrating lymphocytes (S-TILs) had an 80% lower likelihood of disease recurrence compared to those with lower TIL counts, according to a subset analysis of the phase III N9831 trial presented at the 2014 San Antonio Breast Cancer Symposium. The study also found that there was not a similar association between higher S-TIL levels and improved outcomes among patients in N9831 who received standard care with trastuzumab (Herceptin) plus chemotherapy.
Taken together, the findings suggest higher TIL levels may be a biomarker used to identify patients for whom optimal treatment might be chemotherapy alone, without HER2-targeted treatment, suggested Edith Perez, MD, who presented the data at the symposium.
“The message here is that for the patients whose tumors are highly infiltrated with lymphocytes, we cannot prove that there is a benefit to using chemotherapy plus trastuzumab, at least based on the trend that we observed in this study,” said Perez, deputy director at large for the Mayo Clinic Cancer Center.
She cautioned, however, that she is not yet recommending changing clinical practice.
“We need to do additional studies to corroborate our findings. I would not change the standard of care today,” Perez said.
Previous research has suggested a prognostic value for TILs in triple-negative breast cancer (J Clin Oncol
. 2014;32:2959-2966). A study published earlier this year also reported that higher S-TIL levels were linked to a greater trastuzumab benefit among a subset of 209 patients with HER2-positive breast cancer from the FinHER trial (Ann Oncol
. 2014;25(8):1544-1550). The N9831 subset analysis reported at SABCS is the largest analysis to date of the prognostic and predictive value of S-TILs in breast cancer, noted Perez.
The pivotal, three-arm N9831 trial demonstrated improved clinical outcomes when adding trastuzumab sequentially or concurrently to chemotherapy in women with stage I-III HER2-positive breast cancer (J Clin Oncol
The subset analysis that Perez reported included samples from primary tumors of patients enrolled in two of the three N9831 arms: the chemotherapy alone cohort (doxorubicin and cyclophosphamide, followed by paclitaxel) and the concurrent trastuzumab cohort (doxorubicin and cyclophosphamide followed by concurrent paclitaxel plus trastuzumab, and then trastuzumab monotherapy).
Using international consensus guidelines, S-TILs were examined on hematoxylin and eosinstain (H&E) stained whole tumor slides for 489 patients from the chemotherapy-alone group and 456 patients from the concurrent trastuzumab group. Most of the slides were assessed by a single pathologist. Ten percent of cases were read by two pathologists, and a few of those slides were then corroborated by a third pathologist.
Patients with ≥60% S-TIL counts were considered to have high levels (≥60% of cells in the tumor sample were immune cells). Overall, 94 patients met the high S-TIL criterion, including 48 patients in the chemotherapy-alone cohort and 46 patients in the trastuzumab/chemotherapy group.
At a median follow-up of 6.9 years, higher S-TIL levels in the chemotherapy-alone group were associated with a significant recurrence-free survival (RFS) benefit (HR = 0.19; 95% CI, 0.06-0.61; P
= .005). The RFS 10-year Kaplan-Meier projections in this cohort for patients with high and low S-TIL levels were 90.9% and 64.3%, respectively (HR = 0.23; 95% CI, 0.073-0.73; P
“A high level of tumor lymphocytic infiltration was predictive of a benefit with chemotherapy,” Perez said.
In the cohort that received trastuzumab, there was not a link between higher S-TIL levels and improved RFS at the 6.9 year follow-up (HR = 1.01; 95% CI, 0.39-2.6; P
= .98). Kaplan-Meier 10-year RFS estimates for high and low S-TIL patients in this cohort were 80% and 79.6%, respectively (HR = 1.26; 95% CI, 0.5-3.2; P
Perez said she was “surprised” that the trastuzumab results varied from the smaller FinHER analysis and offered possible explanations for the disparity.
“The FinHER study utilized trastuzumab for only 9 weeks. [It was also] a very small study—they only had 49 total events and it was very limited in terms of statistical power,” she said.
Conversely, the N9831 analysis had many more patients and a longer follow-up.
As a separate component of their research, Perez et al also compared the treatment arms head-to-head by S-TIL status. Among patients with high S-TIL levels, there was a nonsignificant trend toward improved RFS in patients receiving chemotherapy alone versus combination therapy with trastuzumab (P
= .22). In patients with low S-TIL levels, there was a significant RFS benefit for patients receiving standard of care with trastuzumab versus chemotherapy alone (P