Article Text

Download PDFPDF

397 Molecular profiling of NSMP high-risk endometrial cancers of the PORTEC-3 trial – prognostic refinement and druggable targets
  1. L Vermij1,
  2. M Powell2,
  3. A Leon-Castillo1,
  4. S De Boer3,
  5. L Mileshkin4,
  6. H Mackay5,
  7. A Leary6,
  8. HW Nijman7,
  9. N Singh8,
  10. P Pollock9,
  11. P Bessette10,
  12. C Haie-Meder11,
  13. V Smit1,
  14. R Edmondson12,
  15. E Crosbie12,
  16. R Nout3,
  17. N Horeweg3,
  18. CL Creutzberg3 and
  19. T Bosse1;13
  1. 1Leiden University Medical Center (LUMC), Pathology, Leiden, Netherlands
  2. 2Barts Health NHS Trust, Clinical Oncology, UK
  3. 3Leiden University Medical Center (LUMC), Radiation Oncology, Leiden, Netherlands
  4. 4Peter MacCallum Cancer Centre, Medical Oncology, Melbourne, Australia
  5. 5Sunnybrook Health Sciences Centre -Odette Cancer Centre, Medical Oncology and Hematology, Toronto, Canada
  6. 6Gustave Roussy, Medical Oncology, Villejuif, France
  7. 7University Medical Center Groningen, Gynaecology, Groningen, Netherlands
  8. 8Barts Health NHS Trust, Pathology, London, UK
  9. 9QUT Gardens Point Campus, Institute of Health and Biomedical Innovation, Brisbane City, Australia
  10. 10Université de Sherbrooke, Obstetrics and Gynaecology, Sherbrooke, Canada
  11. 11Gustave Roussy, Radiation Oncology, Villejuif, France
  12. 12St Mary’s Hospital, Institute of Cancer Sciences, Manchester, UK
  13. 13Leiden University Medical Center, Department of Pathology, Leiden, Netherlands


Introduction/Background*The molecular endometrial cancer (EC) classification has proven prognostic value and can direct adjuvant treatment decisions. Despite this, a relatively large group of EC is still molecularly unclassified (NSMP-EC). In this study we aimed to identify biomarkers among high-risk NSMP-EC patients with prognostic and/or predictive relevance.

Methodology Paraffin-embedded tumour material (n=423) from the PORTEC-3 HREC trial were available for analysis. All patients with NSMP-EC were selected, hence those without pathogenic POLE mutations, mismatch repair deficiency and p53-abnormal immunohistochemistry (IHC). Protein expression of L1CAM, ER and PR (ongoing) were analysed by IHC using a 10% threshold for positivity. Tumour DNA was analysed for pathogenic somatic mutations using a next generation sequencing (NGS) cancer hotspot panel. Time to recurrence was analysed using the Kaplan-Meier method, log-rank tests and Cox’s proportional hazard models.

Result(s)*In total, 126 NSMP-EC were identified in PORTEC-3, the majority were hormone receptor positive (ER n=106/121, 87.6%, PR will be presented at ESGO2021). L1CAM overexpression was observed in 11.2% (n=14/125) and mutations in CTNNB1-exon-3 were identified in 34.3% (n=36/105). Clustering showed that ER-positive NSMP-EC were predominantly endometrioid EC (n=99, 93.4%), low grade (n=88, 83.0%) and L1CAM-negative (n=102, 97.1%) (figure 1). PIK3CA and KRAS mutations were present in 27.3% (n=24) and 19.3% (n=17), respectively. ER-negative NSMP-EC were frequently non-endometrioid (n=11, 73.3%), L1CAM-positive (n=11, 73.3%) and rarely harboured PTEN and CTNNB1 mutations (n=1, 7.1% and 0%, respectively). ER-positivity was associated with lower risk of recurrence (HR 0.32 [95%CI 0.14-0.70]; figure 2A), while L1CAM-overexpression and CTNNB1-exon-3 mutations were not (HR 2.25 [95%CI 0.93-5.43] and HR 1.20 [95%CI 0.57-2.54], respectively). Multivariable analysis confirmed independent favourable prognostic impact of ER-positivity and LVSI. Figure 2B shows impact of LVSI on time to recurrence among patients with ER-positive NSMP-EC.

Abstract 397 Figure 1

Overview of clinicopathological and molecular characteristics of NSMP-EC from PORTEC-3

Abstract 397 Figure 2

Time to recurrence within NSMP-EC

Conclusion*The vast majority of NSMP-HREC are ER-positive (87.6%) and are likely sensitive to hormonal therapy. Other treatment targets might be found in this subgroup too as 27.3% had PIK3CA and 19.3% had KRAS mutations. NSMP-EC with loss of ER-expression were often of non-endometrioid histology and had a high risk of recurrence. Future studies should investigate whether this subgroup would benefit from other systemic therapies.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.