Abstract P5-08-13: Tumor infiltrating lymphocytes and pathological response are prognostic biomarkers in inflammatory and non-inflammatory breast cancer

Background: Tumor-infiltrating lymphocytes (TILs) have been associated with pathologic complete response (pCR) to neoadjuvant chemotherapy (NACT) as well as disease-free (DFS) and overall survival (OS) in certain breast cancer subtypes. pCR has been shown to be predictive of long-term outcome in several neoadjuvant studies and is therefore a potential surrogate marker for patient outcome. The aim of this study was to determine whether TILs and pCR can be used as a prognostic biomarker in inflammatory and non-inflammatory breast cancer.

Materials and Methods: Stromal lymphocytic infiltration (strTILs), defined as the percentage of tumor stroma containing infiltrating lymphocytes (lymphocyte predominant breast cancers (LPBC) cut-off: ≥50%), and pCR, defined as the absence of any residual invasive cancer on the resected breast specimen and all sampled ipsilateral lymph nodes following completion of NACT, were evaluated in 383 (221 Inflammatory (IBC) and 162 non-IBC Locally-advanced (LABC)) breast cancer patients. Tumors were categorised into molecular subtypes and Ki-67 status based on immunohistochemistry. Correlations with clinico-pathological variables, breast cancer-specific (BCSS) and disease-free survival (DFS) were made.

Results: strTILs were present in all patients (median: 15%, IQR: 5% to 30%). There was no difference in the frequency of strTILs between IBC and LABC cases. Thirty three (15%) IBC and 18 (11%) LABC tumors were LPBC. strTILs were significantly more frequent in triple negative (TNBC) (median, 25%) than in HER2+, Ki-67-high (15% for both) and ER/PR+ (10%)(p<0.001; Kruskal-Wallis One Way Analysis of Variance on Ranks). There was a significant association of strTILs with pCR (p<0.001). strTILs median was 27.5%, 15% and 10% for pCR, partial response and no response, respectively (p<0.001). pCR was obtained in 4 (9.1%) of patients with strTILs <10%, in 25 (56.8%) of patients with strTILs between 10 and 40% and in 15 (34.1%) of patients with strTILs >40% (p=1.09E5). strTILs did not predict either DFS or BCSS in the overall breast cancer population. pCR was negatively associated with ER+ (p=0.002), positively with TN (p=0.02) and strongly associated with both DFS & BCSS (p<0.0001, for both). Multivariate analysis showed that, in IBC patients, pCR (p<0.0001) and lymph node rate (p=0.034) were independent predictors for DFS and pCR (p<0.0001), lymph node rate (p=0.034) and LPBC (p=0.024) were independent predictors for BCSS. In LABC, DFS was independently predicted by pCR (p<0.0001) and LPBC (p=0.042) and BCSS by pCR (p<0.0001), LPBC (p=0.005) and ER (p=0.029). LPBC was associated with negative outcome in both IBC and LABC cases.

Conclusion: strTILs showed a strong association with TNBC tumors and with pCR. pCR is a strong prognostic factor for both IBC and LABC. The negative association of LPBC with outcome is unexpected and warrants additional studies.

Citation Format: Kaufman PA, Arias-Pulido H, Colpaert C, Chaher N, Qualls C, Marotti JD, Vermeulen P, Dirix L, van Laere S, Kuppusamy P. Tumor infiltrating lymphocytes and pathological response are prognostic biomarkers in inflammatory and non-inflammatory breast cancer. [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 P5-08-13.

Phase IB study of the combination of docetaxel, gemcitabine, and bevacizumab in patients with advanced or recurrent soft tissue sarcoma: the Axtell regimen

BACKGROUND

To assess the response of patients with soft tissue sarcoma (STS) to the combination of docetaxel, bevacizumab, and gemcitabine. Vascular endothelial growth factor (VEGF)-A levels and expression of VEGF-A and VEGF receptors 1 and 2 were evaluated.

PATIENTS AND METHODS

Thirty-eight chemotherapy-naive patients with STS were enrolled. A dose-finding study for gemcitabine from 1000, 1250, then 1500 mg/m(2) was done in nine patients (three cohorts), followed by an expansion cohort of 27 patients. Dose of docetaxel was 50 mg/m(2), bevacizumab was 5 mg/kg, and gemcitabine was 1500 mg/m(2), every 2 weeks. Serum VEGF-A was measured by enzyme-linked immunosorbent assay and tissue VEGF-A and its receptors by immunohistochemistry.

RESULTS

The median follow-up was 36 months. The overall response rate observed was 31.4%, with 5 complete and 6 partial responses, and 18 stable diseases lasting for a median of 6 months. There was no significant hematologic toxicity. The adverse events with the highest grade were attributed to bevacizumab. There was no correlation of VEGF pathway biomarkers with outcome.

CONCLUSIONS

The combination of gemcitabine, docetaxel, and bevacizumab is safe and effective in patients with STS. The most concerning adverse events were consequences of bevacizumab administration. The benefit of bevacizumab in this patient population remains unclear.

Epidermal growth factor receptor as a biomarker for cervical cancer

This review focuses on the different modes of expression of the epidermal growth factor receptor (EGFR). All methods used to assess EGFR expression are critically analyzed and insights into the use of inhibitors of EGFR for treatment of cervical cancer are discussed. Currently, expression of EGFR as a biomarker for prognosis or for treatment of cervical cancer is not defined for clinical use.

GPR30 and HER-2 Expression in Invasive and Metastatic Breast Cancer

Background: GPR30 expression, a new estrogen receptor, has been previously associated with HER-2, tumor size and metastasis in invasive breast cancer (BC) but its role in paired normal (N), invasive (I), and metastatic (M), samples is unknown. We described GPR30 and HER-2 expression in a collection of paired N/I/M samples, derived from the same individual.Materials and Methods: GPR30 and HER-2 expression was assessed by immunohistochemistry (IHC) in tissue microarrays, containing paraffin-embedded cores from 100 patients diagnosed with invasive BC. N and M samples were also available from the same patient. GPR30 expression was evaluated by an H-score (Intensity (0, negative; 1+, weak; 2+, moderate; 3+, strong) x Percentage of stained epithelial cells). HER-2 expression was evaluated per standard criteria. Log rank tests and Wald tests were employed to assess the clinical impact of these molecular targets on patient outcome based on Kaplan-Meier Product estimator and Cox Proportional Hazard Regression.Results: GPR30 was expressed in 50%, 76%, and 72% of N, I and M, respectively, samples. HER-2 (3+) was found in 14% and 18% of I and M samples, respectively. GPR30 expression in I cases correlated with expression in M cases, and HER-2 expression in I but not M cases. GPR30 expression in M cases correlated with expression in HER-2 expression in M cases. HER-2 expression in I cases correlated with expression in M samples (P<0.05 for all comparisons). GPR30 and HER-2 expression were not associated with grade or stage (P>0.05). GPR30 expression in I or M samples was not associated with either overall survival (OS) or BC-specific survival (BCSS)(P>0.5). HER-2 expression was marginally associated with OS in I (P=0.06; Hazard Ratios (HR): 1.91; 95%CI: 0.956, 3.83) but not in M (P=0.23) cases. HER-2 was significantly associated with BCSS in I cases (P=0.03; HR: 2.39; 95%CI: 1.07, 5.32) and marginally in M (P=0.08) cases.Discussion: A previous study suggested that GPR30 expression predicted the development of metastasis. We found high GPR30 expression in both the primary and metastatic sample but the difference was not significant. While GPR30 expression was not associated with OS or BCSS, HER-2 expression was marginally associated with OS and significantly associated with BCSS in invasive cases.

Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 4158.

Expression of the epidermal growth factor receptor (EGFR) pathway in cervical cancer (CC) patients (pts)

5519

Background: EGFR, a membrane tyrosine kinase receptor that regulates multiple functions such as cell growth, differentiation, gene expression and development through at least 3 pathways, is overexpressed in a wide variety of solid tumors, including CC. The downstream activation of the EGFR pathways has not been studied extensively. The aim of this study is to assess the correlation between EGFR-HER-2-HER-3-HER-4 status, downstream pathways (STAT 3, survivin, RAS) with pt characteristics, overall survival and recurrence free survival in pts diagnosed with invasive CC. Methods: Receptor expressions were assessed by immunohistochemistry on 80 pts from our clinic. Tumors were scored by percentage of cells stained multiplied by intensity for a score range of 0–300 by two pathologists. Median EGFR score was 140. Pts were categorized as low expressor (EGFR score lower than 140) or as high expressor (EGFR score greater than 140). Differences in receptor expression were compared using the Log-rank test for overall (OS) and recurrence-free survivals (RFS) and by Chi-square analyses for the clinical parameters (age, stage, histology, and grade). Results: A total of 80 pts data was analyzed. Mean age of the sample was 48 years. 23% of pts had well differentiated tumors. 80% had squamous histology. 39% had stage IIB or higher. 25% died of cancer. By Log-rank test, standard prognostic factors (age, stage, grade, and histology) showed the expected differences in survival, confirming the validity of the sample. However there were no correlations between the clinical parameters, OS, or RFS with EGFR expression. Data on HER-2–3-4, STAT3, survivin and RAS will be presented at the meeting. Conclusions: Although no correlation has been found between clinical outcome and EGFR expression, EGFR inhibitors of the extracellular domain are expected to be useful for the treatment of CC that overexpress this receptor. These results will serve as baseline data needed to test cetuximab and other EGFR inhibitors in pts with CC.

No significant financial relationships to disclose.