(P 060) TOWARD A NEW GENOMIC AND CLINICO-PATHOLOGIC CLASSIFICATION OF UTERINE ADENOSARCOMAS

Objective: Uterine adenosarcomas (AS) are a rare female malignancy, accounting for only 0.2% of uterine neoplasms. They are characterized by a benign epithelial component associated with a malignant mesenchymal component. They generally have a good prognosis but AS with sarcomatous overgrowth (SO), defined as the presence of pure sarcomatous component in ≥ 25% tumor area, behave more aggressively. The stage of the disease defined by myometrial invasion and extra-uterine extension, and the presence of SO represent most important histological prognostic factors, correlating with the grade of disease. Recently, it was reported that the presence of high-grade morphology in the sarcomatous component, even in the absence of SO, is biologically distinct from low-grade AS and may predict aggressive clinical behavior. Management of localized uterine AS is based on complete surgical removal. Post-operative doxorubicine or platinum-based chemotherapy and radiotherapy can be offered to patients at high risk of relapse (myometrial invasion, high grade, presence of SO). Because of the rarity of uterine AS, management and follow-up guidelines are still lacking.

The aim of our study was to describe clinical and genomic characteristics of a cohort of uterine AS and suggest a new prognostic pathologic classification for uterine AS.

Methods: We analyzed all uterine AS diagnosed at Gustave Roussy between 2010 and 2021. All available material was reviewed by 2 senior pathologists (C.N. and C.G.). For each case, pathological and clinical variables were recorded: tumor size, mitotic rate, presence of SO and heterologous component, age, location, FIGO stage, surgery, chemotherapy regimens, radiotherapy or brachytherapy, recurrence and status at last FU. High grade morphology was defined as severe nuclear atypia with pleomorphism. Overall survival (OS) was defined as the time from treatment initiation until either the date of death, the date of last follow-up or the date of end of analysis. Progression-free survival (PFS) was defined as the time from treatment initiation until disease progression or death, whichever occurred first. We subsequently investigated genome-wide copy number alterations (CNA) with the Affymetrix OncoScan™ CNV Assay. All statistical analyses were performed using SAS software.

Results: 67 cases of AS with available histologic material were included. We identified 19/67 (28%) patients with low grade AS (LGAS), 6/67 (9%) LGAS with SO (LGASSO), 8/67 (12%) high grade AS (HGAS) and 34/67 (51%) HGAS with SO (HGASSO).

Clinical and follow up data were available for 53/67 (79.1%) patients.
Symptoms leading to the diagnosis of AS were mainly metrorrhagia and abdominal pain.
Median age was 59 years [range 16-87]. At diagnosis, 23/53 (43.3%) patients had tumor FIGO stage IA, 18 stage IB, 2 stage IC, 5 stage II, 3 stage III, 2 stage IV. 52/53 patients had a macroscopically complete surgery, 1 patient had an incomplete surgery.
20/53 (37.7%) patients received post-operative chemotherapy, among them 11 had a stage I high grade AS, 6/53 (49%) patients received radiotherapy and 21/53 (39.6%) brachytherapy.
13/53 (24.5%) patients relapsed: 9 HGASSO, 1 HGAS with rhabdomyoblasts, 3 LGASSO. Among them, 4 patients didn’t receive post-operative therapy, 7 had received only one treatment modality and 2 pts had received all 3 treatment modalities.

Median OS and PFS were not reached. OS rates at 5 years (95% CI) were : 100% [100% ; 100%] for HGAS, 70.6% [55.3% ; 90%] for HGASSO, 100% [100% ; 100%] for LGAS and 66.7% [30% ; 100%] for LGASSO (Figure 1).
PFS rates at 5 years (95% CI) were :75% [42.6% ; 100%] for HGAS, 53.7% [36.7% ; 78.6%] for HGASSO, 100% [100% ; 100%] for LGAS and 75% [42.5% ; 100%] for LGASSO (Figure 2).

Presence of SO, regardless of histologic grade was significantly associated with worse survival (AS vs ASSO : p=0,029) and worse recurrence rates (AS vs ASSO : p=0,013) (Figure 3-4).

Genomewide CNAs was investigated on 33/67 (49%) cases. LGASSO samples displayed only few CNAs with some aberrations in common with HGAS. Most of LGAS samples revealed a balanced genome (Results presented by C. Ngo at USCAP 2022: Uterine Adenosarcoma: Histologic Reclassification of 67 Cases with Clinical Correlation and CGH-array Analysis).

Conclusion: To our knowledge, this is one of the largest retrospective cohorts of uterine AS. Histopathological classification of AS into 4 groups, taking into account the morphologic grade and the presence of SO might predict clinical behavior and correlates with the genomic copy number landscape. Presence of SO, regardless of histologic grade was significantly associated with recurrence. Despite being very rare, LGASSO is clinically more aggressive than LGAS and share some common chromosomal losses as HGAS. Close and long-term follow up should therefore be proposed to patients with LGASSO.

Figure 1.docx

Figure 2.docx

Figure 3.docx

Figure 4.docx