Association of Genomic Instability Score, Tumor Mutational Burden, and Tumor-Infiltrating Lymphocytes as Biomarkers in Uterine Serous Carcinoma

Abstract

Background: Uterine serous carcinomas represent 10% of uterine carcinomas but account for nearly 40% of deaths from the disease. Improved molecular characterization of these tumors is instrumental in guiding targeted treatment and improving outcomes. This study assessed the genomic instability score (GIS), tumor mutational burden (TMB), and tumor-infiltrating lymphocytes (TILs) in patients with USC. Methods: A retrospective cohort study evaluated patients with USC following staging surgery. The GIS and TMB were determined from archived specimens. We evaluated the tumoral expression of CD3, CD4, CD8, FOXP3, and CD68 using immunohistochemistry. T-tests were used to assess associations of TILs with the GIS. Results: We evaluated 53 patients with USC. The median GIS was 31 (range: 0−52) and a higher GIS was not associated with progression-free (PFS) or overall survival (OS). The median TMB was 1.35 mt/Mb; patients with TMB > 1.35 mt/Mb had improved PFS and OS (p = 0.005; p = 0.002, respectively). Tumors with increased CD3+ and CD4+ immune cells had a higher mean GIS (p = 0.013, p = 0.002). Conclusions: TMB > 1.35 mt/Mb was associated with improved survival in USC patients, whereas the GIS was not. Lower TMB thresholds may provide prognostic value for less immunogenic tumors such as USC. In this limited cohort, we observed that increased TIL populations were correlated with a higher GIS.

Keywords: genomic instability; immunohistochemistry; tumor mutational burden; tumor-infiltrating lymphocytes; uterine serous cancer.

Figure 1

Frequency of mutations in genes associated with molecular classification and other biomarkers (TP53, POLE, ERBB2, and MMR genes) and genes associated with the HRD pathway (BRCA1, BRCA2, PALB2, RAD51C, RAD51D, BRIP1, and BARD1) in uterine serous cancers. This analysis includes point mutations, insertions, deletions, and rearrangements. Unknown mutations were mutations with unknown pathogenic significance. Abbreviations: HRD, homologous recombination deficiency.

Figure 2

Representative examples of immune cell infiltration for CD3, CD4, CD3, FOXP3, and CD68. The upper row represents high immune cell infiltration (>75th percentile). The lower row represents low–normal immune cell infiltration (≤75th percentile). Scale bar is 0.03 mm.

Figure 3

(A–C): Distribution lymphocytes by genomic instability score (GIS) and tumor recurrence; pink = recurrent, black = no recurrence. Quadrants defined by GIS > 31 (median) or ≤31 and number of TILs > or ≤the 75th percentile. (D–F): Distribution lymphocytes by tumor mutational burden (TMB) and tumor recurrence. Quadrants defined by TMB > 1.35 (median) or ≤1.35 and number of TILs > or ≤the 75th percentile. (A): CD3+ lymphocytes, 75th percentile is 76.8, (B): CD4+ lymphocytes, 75th percentile is 28.8, (C): CD8+ lymphocytes, 75th percentile is 28.8.

Figure 4

Comparison of mean GIS score by high (H) and low–normal (L–N) immune cell counts. (A): CD3, (B): CD4, (C): CD8. Abbreviations: GIS, genomic instability score.

Figure 5

Survival outcomes for patients with uterine serous cancer by TMB and GIS. (A,B): PFS and OS with TMB above and below or equal to the median, 1.35. (C,D): PFS and OS for patients with GIS above and below or equal to the median. Abbreviations: GIS, genomic instability score; TMB, tumor mutation burden; PFS, progression-free survival; OS, overall survival.