To the Editor:
The editorial by Dr. Robert Norman in this issue of Fertility and Sterility (1) succinctly summarizes the various approaches currently used to identify biomarkers of endometrial receptivity toward development of a viable predictive/assessment test for clinical application. As Dr. Norman points out, multiple approaches have been applied to identify candidate biomarkers, including proteomic and lipidomic assessments of uterine aspirate fluid and gene expression profiling of endometrial biopsies. In this editorial, Dr. Norman introduces our paper in which we demonstrate the feasibility of performing genome-wide gene expression profiling on uterine aspirations (2). Using unsupervised hierarchical clustering, we demonstrate that the phase of sampling (LH+2 versus LH+7) affects gene expression more so than individual differences in gene expression between patients. We identified and verified robust differences in expression of 245 genes due to phase of sampling and determined that expression of 53 of these genes efficiently separated our two groups and those of a publically available dataset of gene expression signatures obtained from endometrial biopsy samples. Since LH+2 coincides with a pre-receptive phase and LH+7 coincides with a receptive phase in fertile women, the differentially expressed genes identified in our study encode for candidate biomarkers of endometrial receptivity. This 53-gene list overlaps significantly with those indicated in other studies, including the 238-gene list comprising the endometrial receptivity assay used by Diaz-Gimeno et al. (3) in their algorithm for predicting the receptive period from endometrial biopsy samples. We are excited by the potential of our approach because uterine aspiration is less traumatic to the endometrium than biopsy. Unlike endometrial biopsy, uterine aspiration is compatible with evaluation within an active IVF or natural cycle in which a patient is attempting pregnancy (4, 5), enabling us and other researchers to directly associate altered gene expression with implantation success. This approach will facilitate our future identification of those candidate biomarkers for further development of point-of-care assays for clinical use. While our 53-gene cassette includes several interesting candidate biomarkers, including those encoding secreted products, we do not propose these as a clinical test. Rather, we propose that our approach enables further testing and reduction of this signature to identify those genes most predictive. We agree with Dr. Norman that protein-based assays will be preferable to transcript measurements for clinical assays; however, gene expression data are necessary to direct proteomic discovery assays to improve these approaches.
Dr. Norman points out that we did not fully characterize the cell types present in the uterine aspirations and that changes in cell types would impact gene expression signatures. However, this criticism applies equally to proteomic approaches and to gene expression studies of endometrial biopsies and misses the point. Alterations in infiltrating immune cells undoubtedly contribute to the gene expression profiles just as they contribute to differences in proteins present in the aspirated fluid and likely impact on receptivity of the endogenous intrauterine environment. Our discovery approach mirrors the situation one would encounter in a clinical testing situation, with multiple cell types contributing to the proteome or transcriptome. Dr. Norman also raises an issue of blood contamination and an inability to obtain uterine fluid from patients. We did not encounter either of these issues. Prior to adopting the technique, we trialed various catheters in patients undergoing surgery. We selected the catheter used in our study as it was cost-effective, reliably acquired fluid and cells, did not inflict pain, and only very few red blood cells were noted on histology as reviewed by a pathologist.
Lastly, Dr. Norman seems to suggest that we are proposing our 53-gene signature as a clinical test for endometrial assessment. It is important to stress that this is not our claim in this paper. Rather, we recommend adoption of our less-invasive sampling approach, which will allow the further validation of gene expression changes associated with the clinically receptive period. We point out that this approach can be used to directly associate biomarker expression with implantation success, and that these findings can guide proteomic discovery and assay development. Such studies are presently underway in our laboratory.
Crystal Chan, M.D.a,b,c,d, Theodore J. Brown, Ph.D.a,b,c, and Ellen M. Greenblatt, M.D.a,d
a. Department of Obstetrics and Gynaecology, and b. Institute of Medical Sciences, University of Toronto;
c. Samuel Lunenfeld Research Institute and d. Centre for Reproductive Health and Fertility, Mount Sinai Hospital, Toronto, Ontario, Canada
1. Norman RJ. Biomarkers of endometrial receptivity through a minimally invasive approach. Fertil Steril 2013.
2. Chan C, Virtanen C, Winegarden NA, Colgan TJ, Brown TJ, Greenblatt EM. Discovery of biomarkers of endometrial receptivity through a minimally-invasive approach: a validation study with implications for assisted reproduction. Fertil Steril 2013.
3. Diaz-Gimeno P, Ruiz-Alonso M, Blesa D, Bosch N, Martinez-Conejero JA, Alama P, et al. The accuracy and reproducibility of the endometrial receptivity array is superior to histology as a diagnostic method for endometrial receptivity. Fertil Steril 2013;99:508-17.
4. Boomsma CM, Kavelaars A, Eijkemans MJ, Amarouchi K, Teklenburg G, Gutknecht D, et al. Cytokine profiling in endometrial secretions: a non-invasive window on endometrial receptivity. Reprod Biomed Online 2009;18:85-94.
5. van der Gaast MH, Beier-Hellwig K, Fauser BC, Beier HM, Macklon NS. Endometrial secretion aspiration prior to embryo transfer does not reduce implantation rates. Reprod Biomed Online 2003;7:105-9.
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2013.05.051