TILs Associated With Survival for Advanced HER2+ Breast Cancer

Stephen J. Luen, MBChB

Higher levels of tumor-infiltrating lymphocytes (TILs) were associated with improvements in overall survival (OS) for patients with advanced HER2-positive breast cancer treated with docetaxel, trastuzumab (Herceptin), and pertuzumab (Perjeta) in the phase III CLEOPATRA trial, according to findings presented at the 2016 San Antonio Breast Cancer Symposium.¹

In the study, which was simultaneously published in Lancet Oncology,² each 10% increase in stromal TILs was associated with an 11% improvement in OS by multivariate analysis, said lead investigator Stephen J. Luen, MBChB (adjusted HR, 0.89; 95% CI, 0.83-0.96, P = .0014). Conversely, however, TILs were not significantly connected with progression-free survival (PFS; adjusted HR, 0.95; 95% CI, 0.90-1.00; P = .063).

"This is the first study investigating associations between TILs and survival in advanced HER2-positive breast cancer in the context of first-line treatment with dual HER2 blockade with trastuzumab and pertuzumab," said Luen, from the Peter MacCallum Cancer Centre, University of Melbourne, Melbourne, VIC, Australia. "We found a nonsignificant trend between higher TIL levels and PFS. On the other hand, there was a significant association between higher TILs and reduced risk of death. This effect was linear."

The phase III CLEOPATRA trial randomized 808 patients with previously untreated HER2-positive advanced breast cancer to receive docetaxel and trastuzumab with pertuzumab (n = 402) or without placebo (n = 406). At a median follow-up of 50 months, median PFS was improved by 6.3 months and the median OS was extended by 15.7 months with the addition of pertuzumab (P <.001).

"This has set the standard of care as first-line therapy for advanced HER2-positive breast cancer," said Luen. "To determine the prognostic association between stromal TILs and PFS in patients with advanced HER2-positive breast cancer treated in the first-line setting. This is becoming increasingly relevant with the emergence of immunotherapies and their potential to improve survival for breast cancer patients."

At entry into the CLEOPATRA trial, tissue samples were prospectively collected for later analyses, with 678 evaluable for TIL assessment (84%). Twenty-three percent of the samples were freshly obtained ≤45 days from the time of randomization from patients who had not received prior endocrine therapy (n = 155). The remaining samples were archival (77%; n = 519).

ER-negative tumors had significantly higher levels of TILs compared with ER-positive disease (15% vs 10% of stroma; P <.001). TIL levels also differed by ethnicity (P <.001). There were more TILs in the stroma for Asians (15%) versus Caucasians (10%; P = .002) and African Americans or blacks (5%). Additionally, more TILs were observed in archival samples versus fresh tissue (15% vs 10%; P <.001).

In a small sample of 40 primary (n = 20) and metastatic samples (n = 20), there were fewer TILs in the metastatic group. Overall, the stroma consisted of 10% TILs for the primary samples versus 7.5% for the metastatic tissue, although this was not statistically significant (P = .07).

The number of TILs seen in metastatic tissue biopsies was generally lower, except for those to the lung. Metastases to the bone (n = 9), liver (n = 20), and skin (n = 23) had TIL levels of 5% compared with a 30% level for those with lung metastases (n = 6; P <.001).

"I would express a bit of caution, because we only had a small number of samples," Luen said, regarding the differences in TIL count by metastatic site. "With the lung, it was interesting that it was higher. There have been anecdotal reports of response to checkpoint blockade in patients with limited burden of disease to just the nodes or lung. I think this is hypothesis generating and needs to be confirmed in other studies."

In the multivariate assessment, the overall number of stromal TILs was the only statistically significant factor associated with survival. This effect was linear, noted Luen. Looking at the categories that differed by TIL levels, the OS hazard ratio for ER-positive versus -negative tumors was 0.79; however, this was not statistically significant (95% CI, 0.60-1.04; P = .09).

The median OS was not reached for those with TILs >20% treated with pertuzumab. The 3-year OS rate was 78%. For those with TILs ≤20% in the pertuzumab arm, the median OS was 49.1 months and the 3-year OS rate was 64%. In the placebo arm, the median OS was 41.2 and 36.6 months for those with TILs >20% and ≤20%, respectively. The 3-year OS rates were 55% and 50% in the >20% and ≤20% placebo groups, respectively.

There was an association between TILs and survival by multivariate analysis for those who received pertuzumab compared with placebo (HR, 0.66; 95% CI, 0.51-0.87; P = .003). However, the levels of TILs were not predictive of benefit for both PFS (P = .23) and OS (P = .21). "There was no predictive effect observed with regard to pertuzumab treatment," said Luen.

"The TILs had a strong prognostic effect in the pertuzumab arm; however, the interaction P value for this was not significant for PFS and OS," he added. "Importantly, patients derived benefit from pertuzumab irrespective of TIL levels and so TIL levels should not be used to determine who should not receive pertuzumab treatment."

References:

1. Luen SJ, Salgado R, Fox S, et al. Prognostic associations of tumor-infiltrating lymphocytes (TIL) in metastatic HER2-positive breast cancer (BC) treated with trastuzumab and pertuzumab: A secondary analysis of the CLEOPATRA study. Presented at: San Antonio Breast Cancer Symposium; San Antonio, TX; December 6-10, 2016. Presentation S1-08.
2. Luen SJ, Salgado R, Fox S, et al. Tumour-infiltrating lymphocytes in advanced HER2-positive breast cancer treated with pertuzumab or placebo in addition to trastuzumab and docetaxel: a retrospective analysis of the CLEOPATRA study [Published online December 7, 2016]. Lancet Oncol. 2016.

<<<

View more from the 2016 San Antonio Breast Cancer Symposium

The median age of patients in the TIL evaluable cohort was 54 years (range, 22-89) and they were primarily Caucasian (58%; n = 391), Asian (34%; n = 230), and African American or black (4%; n = 24). Forty-eight percent of patients had ER-positive disease and 22% had a mutated PIK3CA gene. Neoadjuvant or adjuvant therapy was received by 42% of patients, with 11% having received neoadjuvant or adjuvant trastuzumab before study entry. Seventy-nine percent of patients had visceral disease at screening.