Elevation of TP53 Autoantibody Before CA125 in Preclinical Invasive Epithelial Ovarian Cancer

Abstract

Purpose: The TP53 tumor-suppressor gene is mutated in >95% of high-grade serous ovarian cancers. Detecting an autologous antibody response to TP53 that might improve early detection.Experimental Design: An immunoassay was developed to measure TP53 autoantibody in sera from 378 cases of invasive epithelial ovarian cancer and 944 age-matched healthy controls from the United States, Australia, and the United Kingdom. Serial preclinical samples from cases and controls were also assayed from the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS).Results: Using a cutoff value of 78 U/mL to achieve a specificity of 97.4%, TP53 autoantibody was elevated in 30% of 50 cases from MD Anderson, 21.3% of 108 cases from the Australian Ovarian Cancer Study, and 21% of 220 cases from the UKCTOCS. Among 164 cases with rising CA125 detected with the UKCTOCS risk of ovarian cancer algorithm (ROCA), 20.7% had elevated TP53 autoantibody. In cases missed by the ROCA, 16% of cases had elevated TP53 autoantibody. Of the 34 ovarian cancer cases detected with the ROCA, TP53 autoantibody titers were elevated 11.0 months before CA125. In the 9 cases missed by the ROCA, TP53 autoantibody was elevated 22.9 months before cancer diagnosis. Similar sensitivity was obtained using assays with specific mutant and wild-type TP53.Conclusions: TP53 autoantibody levels provide a biomarker with clinically significant lead time over elevation of CA125 or an elevated ROCA value. Quantitative assessment of autoantibodies in combination with CA125 holds promise for earlier detection of invasive epithelial ovarian cancer. Clin Cancer Res; 23(19); 5912-22. ©2017 AACR.

Conflict of interest statement

Disclosure of potential conflicts of interest: Dr. Wei-Wu He is the former chief executive officer of OriGene Technologies, Inc. Dr. Robert Bast receives royalties from Fujirebio Diagnostics Inc. for discovery of CA125. Other authors have no conflicts of interest.

©2017 American Association for Cancer Research.

Figures

Figure 1

TP53 autoantibody titers and CA125 values in serum samples from the MDACC-NROSS study. A, Box plots for TP53 autoantibody and CA125. Each box exhibits maximum, upper quartile, median, lower quartile and minimum values. Triangles represent cases and circles represent controls. The lower tables display TP53 autoantibody and CA125 positive numbers and percentages in the case and control groups. B, Scatter plot of TP53 autoantibody titers and CA125 values. Triangles represent ovarian cancer cases and circles represent controls. The black dashed line represents the common cut-off value (78 U/mL) of TP53 autoantibody.

Figure 2

TP53 autoantibody titers in serum samples from the AOCS biobank data set. Box plots for TP53 autoantibody titers. Each box exhibits maximum, upper quartile, median, lower quartile and minimum values. Triangle symbols represent cases, square symbols represent benign ovarian neoplasms, and circle symbols represent controls. The lower table displays TP53 autoantibody positive (above common cut-off 78 U/mL) numbers and percentages in case, benign, and control groups.

Figure 3

TP53 autoantibody titers and CA125 values in serum samples from the UKCTOCS screening trial. A, Box plots for TP53 autoantibody and CA125. Each box exhibits maximum, upper quartile, median, lower quartile and minimum values. Triangles represent cases and circles represent controls. The lower table displays TP53 autoantibody positive numbers and percentages in ROCA/CA125 (+) case, ROCA/CA125 (-) case and control groups. B, Scatter plots of TP53 autoantibody levels and CA125 values. Triangles represent ovarian cancer cases and circles represent controls. The black dashed line represents the common cut-off value (78 U/mL) for TP53 autoantibody.

Figure 4

Longitudinal analysis of CA125 values and TP53 autoantibody titers in pre-diagnostic serial serum samples from ovarian cancer patients in the UKCTOCS study. A, Lead time prior to diagnosis for elevated TP53 autoantibody, ROCA and CA125 (>35 U/mL) in 43 cases with elevated TP53 autoantibody. In the left panel solid lines were detected with the ROCA and the dashed lines were not. In the middle panel, circles were detected with the ROCA and the triangles were not. B, Lead time prior to diagnosis for elevated TP53 autoantibody, ROCA and CA125 (>35 U/mL) in 34 cases with elevated TP53 autoantibody who were diagnosed with the ROCA. C, Lead time prior to diagnosis for elevated TP53 autoantibody, ROCA and CA125 (>35 U/mL) in 9 cases with elevated TP53 autoantibody who were not diagnosed with the ROCA. Sample pair number indicates the index of a given sample pairing within each set, ranging from 1 to 46 in A, 1 to 37 in B, and 1 to 9 in C. Colors match between the two plots in each panel, so the dramatic downward segment in C is seen to correspond to sample pair 2, with a difference in detection times of around 80 months.

Figure 5

ROC curve analysis for TP53 autoantibody and CA125 biomarkers in the UKCTOCS study. A, The AUC, partial AUC (pAUC) and sensitivity for TP53 autoantibody, CA125 and the combination of these two markers are shown in the table (p value = 0.001 for AUC of CA125 vs CA125+TP53 AAb; p value = 0.097 for pAUC of CA125 vs CA125+TP53 AAb). Dotted line, ROC curve for TP53 autoantibody; dashed line, ROC curve for CA125; solid line, ROC curve for the combination of two biomarkers. The gray vertical lines are corresponding to 0.98 specificity. B, ROC curves for TP53 autoantibody and CA125 in the UKCTOCS study were calculated when 3 months was chosen as the cut-off to define the “significant early detection”. The AUC, partial AUC (pAUC) and sensitivity for TP53 autoantibody, CA125 and the combination of these two markers are shown in the table (p value = 0.000 for AUC of CA125 vs CA125+TP53 AAb; p value = 0.983 for pAUC of CA125 vs CA125+TP53 AAb).