Study Finds Analyzing DNA From Pap Smears Could Help Diagnose Ovarian Cancer Early

Findings from a small study published in Science Translational Medicine suggest DNA taken from Papanicolaou (Pap) smears could potentially help detect ovarian cancer early. Ovarian cancer is often diagnosed at a late stage in part due to a lack of early detection tests, but the study, conducted in Italy, provides proof of principle for an early ovarian cancer test (EVA) based on high copy number profile abnormality (CPA) found in DNA from Pap smears.

High-grade serous ovarian cancer (HGSOC) is the gynecological cancer with the highest mortality rate, with this trend driven by late diagnosis due to non-specific symptoms in early stages and a lack of early screening methods, the study authors explained. Therefore, they aimed to determine whether DNA derived from samples taken during routine Pap smears in presymptomatic women could be used to diagnose HGSOC earlier.

“In the work described here, we focused on 2 early molecular events occurring during HGSOC development: TP53 mutations and genomic instability,” the authors wrote. “The results presented here are consistent with recent evidence from large-scale pan-cancer studies suggesting that in many cancer types, genomic alterations are present before the disease is detectable and may possibly contribute to disease development.”

The investigators used a retrospective, multicentric selection of 250 archival routine Pap smear samples from 113 women who were presymptomatic at the time of collection but were later diagnosed with HGSOC, as well as samples from 77 healthy women. Samples were from between 1 month and 13.5 years before diagnosis with HGSOC. Low-pass whole-genome sequencing (WGS) of DNA from the Pap samples was used to detect genome instability in terms of CPA.

DNA mutation | Image credit: nobeastsofierce - stock.adobe.com

Somatic copy number alterations (SCNAs) are common in cancer but rarely found in normal tissue, and therefore were the marker of interest in the study. The CPA score used in the EVA test reflects the overall amount of SCNA in the tumor genome instead of the amount in single genomic regions.

The samples from women later diagnosed with HGSOC were divided into 2 groups: A (n = 62) and B (n = 51). The EVA test was developed on cases enrolled in group A based on genome-wide copy number instability as a biomarker for early progression, and those in group B were used to validate the EVA based on clonal pathogenic TP53 somatic variants. The median age at diagnosis was 60 years in group A (range, 40-81) and 61 years in group B (range, 42-80). The median age of healthy women was 43 years (range, 20-64), and the authors noted that procuring Pap test smears from healthy women at a similar age to groups A and B was difficult because Pap test analysis is no longer recommended in postmenopausal women.

CPA values in DNA from Pap smears from groups A and B were higher compared with Pap smears from healthy women. Investigators were able to stratify CPA values into 3 intervals when testing both a region-specific approach based on SCNA reported in the ovarian cohort of the Cancer Genome Atlas Ovarian Cancer (TCGA-OV) as well as a genome-wide approach.

“The use of a numerical measure implies the selection of acceptable thresholds to maximize the number of true positives and minimize or eliminate, if possible, the number of false positives,” the authors wrote. “To this aim, we defined two CPA thresholds that include a ‘gray zone’ in which the presence of genomic alterations is considered uncertain. Furthermore, we adopted a conservative approach to ensure high confidence in calling a sample positive for genomic alterations at the cost of registering more false negatives.”

With the genome-wide approach, the intervals identified were classified as negative for genomic alterations (CPA between 0 and 0.52372), a “gray zone” or area of uncertainty (CPA between 0.52372 and 0.61), and positive for genomic alterations (CPA of 0.61 or greater). The specificity of the EVA was 96.0% (95% CI, 88.35-100.0), sensitivity was 75.38% (95% CI, 64.97-85.79), and accuracy was 81.11%.

“We surmise that this sensitivity is satisfactory, considering that the archival Pap test collection was not originally meant to be used for DNA analysis, which implies that the sampling procedure and storage conditions were probably suboptimal for the aims of this study,” the authors wrote.

The study was limited by its retrospective nature and limited sample size, lack of standardized time points for collection, and difference in the median ages of healthy women vs groups A and B. The authors noted that SCNA is known to accumulate with age. Constitutive copy numbers were removed from the EVA assay to minimize the impact of these genomic variations, they added.

“Our study provides the basis for a new approach to the early detection of HGSOC based on the assessment of genomic instability patterns of DNA extracted from cervical smears,” the authors concluded. “Because the low survival of patients with HGSOC is usually related to the delay in diagnosis, we believe that the application of the approach proposed here may have a marked impact on mortality from this neoplasm.”

Reference

Paracchini L, Mannarino L, Romualdi C, et al. Genomic instability analysis in DNA from Papanicolaou test provides proof-of-principle early diagnosis of high-grade serous ovarian cancer. Sci Transl Med. 2023;15(725):eadi2556. doi:10.1126/scitranslmed.adi2556