The effect of pre-analytical variables on downstream application and data analysis of human endometrial biopsies

What are the effects of pre-analytical variables on the downstream analysis of patient-derived endometrial biopsies?

There are distinct differences in the protein levels of the master regulator of oxygen homeostasis, hypoxia-inducible factor-1-alpha (HIF1α), and the protein and mRNA levels of three related genes, carbonic anhydrase 9 (CA9), vascular endothelial growth factor A (VEGFA) and progesterone receptor (PR) in human endometrial biopsies, depending on the pre-analytical variables: disease status (cancer vs benign), timing of biopsy (pre- vs post-hysterectomy) and type of biopsy (pipelle vs full-thickness).

Patient-derived biopsies are vital to endometrial research, but pre-analytical variables relating to their collection may affect downstream analysis, as is evident in other tissues.

A prospective observational study including patients undergoing hysterectomy for endometrial cancer (EC) or benign indications was conducted at a large tertiary gynaecological unit in the UK. Endometrial biopsies were obtained at different time points (pre- or post-hysterectomy) using either a pipelle endometrial sampler or as a full-thickness wedge biopsy.

The changes in HIF1α, CA9, VEGFA and PR protein levels were measured by semi-quantitative analysis of immunostaining, and the expression levels of three genes (CA9, VEGFA and PR) were investigated by quantitative real-time PCR, in endometrial biopsies from 43 patients undergoing hysterectomy for EC (n = 22) or benign gynaecological indications (n = 21).

An increase in HIF1α immunostaining was observed in EC versus benign endometrium (functionalis glands) obtained pre-hysterectomy (P < 0.001). An increase in CA9 immunostaining was observed in EC versus benign endometrial functionalis glands at both pre- and post-hysterectomy time points (P = 0.03 and P = 0.003, respectively). Compared with benign endometrial pipelle samples, EC samples demonstrated increased mRNA expression of CA9 (pre-hysterectomy P < 0.001, post-hysterectomy P = 0.008) and VEGFA (pre-hysterectomy P = 0.004, post-hysterectomy P = 0.002). In benign uteri, HIF1α immunoscores (functionalis glands, P = 0.03 and stroma, P = 0.009), VEGFA immunoscores (functionalis glands, P = 0.03 and stroma, P = 0.01) and VEGFA mRNA levels (P = 0.008) were increased in matched post-hysterectomy versus pre-hysterectomy samples. Similarly, in EC, an increase in VEGFA immunoscores (epithelial and stromal) and VEGFA mRNA expression was observed in the matched post-hysterectomy versus pre-hysterectomy biopsies (P = 0.008, P = 0.004 and P = 0.018, respectively). Full-thickness benign post-hysterectomy endometrial biopsies displayed increased VEGFA (P = 0.011) and PR (P = 0.006) mRNA expression compared with time-matched pipelle biopsies.

N/A.

This descriptive study explores the effect of pre-analytical variables on the expression of four proteins and three hypoxia-related genes in a limited number of endometrial biopsies from patients with EC and benign controls. Due to the small number, it was not possible to investigate other potential variables such as menstrual cycle phase, region-specific differences within the endometrium, grade and stage of cancer, and surgical technicalities.

Careful consideration of the effects of these pre-analytical variables is essential when interpreting data relating to human endometrial biopsies. A standardized approach to endometrial tissue collection is essential to ensure accurate and clinically transferrable data.

The authors have no conflicts of interest to declare. The work included in this manuscript was funded by Wellbeing of Women project grants RG1073 and RG2137 (D.K.H.), Wellbeing of Women Entry-Level Scholarship ELS706 and Medical Research Council MR/V007238/1 (A.M./D.K.H.), Liverpool Women's Hospital Cancer Charity (M.A.) and University of Liverpool (L.B., L.R. and E.N.).