Open questions — uterine adenosarcoma¶
Last curated: 2026-10-31. In this condition, “open” usually means no adequately powered study exists, not that two well-powered studies disagree. A question that can only be answered by a rare-tumour consortium or pathology-linked registry says so. Stable IDs (OQ-n). Supporting PMIDs/NCTs from the 2026-08-31 build were re-checked in later sessions; 2026-10-10 and 2026-10-31 additions are live-resolved.
How to read priority¶
Tier 1 = answering it would change practice or reorder counselling, and a study is designable today (usually as a multi-institutional registry or basket, not a classical RCT). Tier 2 = important, but blocked on tools, numbers, or a Tier-1 answer. Tiers are curator judgment.
Dots not yet connected¶
Cross-domain junctions where two bodies of evidence both exist in this knowledge base but no study has joined them.
| # | Dot A | Dot B | The missing junction | Powers |
|---|---|---|---|---|
| D1 | Morcellation-safety literature is written about uterine sarcoma as a class (occult sarcoma 1/278 leiomyoma hysterectomies; PMID 26646120, 30447212) | Adenosarcoma-specific presentation as a polyp, not a fibroid, and SO-stratified outcome (PMID 2156771, 25449308) | No study reports how often adenosarcoma is morcellated, or whether morcellation changes adenosarcoma-specific survival | OQ-3 |
| D2 | SO is a morphological ≥25% volume call (PMID 2535774) with independent survival effect (PMID 25449308) | SO tracks copy-number complexity, not mutation count (mean CNVs 24.6 vs 5; PMID 25231023); high-grade disease is TP53-enriched (PMID 28834809) | No study has asked whether a genomic complexity score outperforms or refines the 25% volume rule | OQ-1, OQ-4 |
| D3 | PI3K/AKT/PTEN pathway alteration in 13/18 (72%) in a selected sequencing series (PMID 25231023) | No PI3K-pathway-selected adenosarcoma trial identified (ClinicalTrials.gov search rerun 2026-10-31) | The most frequent pathway finding in that small series has not yet been tested prospectively in this histology | OQ-6 |
| D4 | Carroll’s non-significant adjuvant signal in stage I SO (PFS 46.7 vs 29.4 months, p=0.28; PMID 25449308) | ESGO “option” vs French “not established” for adjuvant chemo (PMID 39322612, 37202293) | Guidelines disagree in wording over a comparison that has never been run multi-institutionally | OQ-2 |
| D5 | Nodal metastasis ~3% but aHR 2.34 for CSS when present (PMID 28109626) | SO, large tumour and deep invasion associated with nodal risk in a 230-case literature review (PMID 28109626) | No prospective selection rule for lymphadenectomy has been tested | OQ-5 |
| D6 | Late local recurrence (⅓ of Clement recurrences ≥5 years; PMID 2156771) and lung as 75% of Tate’s distant recurrences (PMID 29441675) | No prospective surveillance schedule | Timing and site of imaging are extrapolated, not measured against outcome | OQ-7 |
| D7 | den Hollander HRQoL in mixed uterine sarcoma (n=13): sexual/urological/menopausal, shock, missing sarcoma knowledge (PMID 35443673) | Standard operation is TAH-BSO (PMID 27718181) with ovarian preservation unproven in this histology (PMID 28541635) | No adenosarcoma-specific HRQoL module or ovarian-preservation decision aid exists | OQ-8 |
| D8 | ATRX mutation in 2 high-grade recurrences after low-grade primaries (PMID 36138078) | Pathology-at-recurrence influences OS (PMID 29445312) | Low-grade → high-grade transformation is observed, not mapped as a surveillance or treatment trigger | OQ-4, OQ-7 |
| D9 | Atypical polyps share 12q13–15 and 6q25.1 gains with early adenosarcoma but had zero progression over 150 months (PMID 34675347) | Diagnostic criteria versus adenofibroma remain mitotic-count conventions from 1981/1990 (PMID 6263458, 2156771) | Morphology, copy number and outcome have not been jointly modelled as a diagnostic rule | OQ-9 |
When a junction study appears, connect the dot here and sharpen or close the powered question.
Tier 1 — highest-value directions¶
OQ-1. Is anything other than sarcomatous overgrowth independently prognostic once analyses are restricted to multivariable models with sarcoma-pathologist review?¶
SO is the consistent independent factor (Carroll 2014, PMID 25449308; Howitt 2015, PMID 25231023; Aytekin 2026, PMID 42794871). LVSI is the leading additional candidate (median OS 1.0 vs 8.9 years; Nathenson 2018, PMID 30044322). Aytekin’s seven-centre n=43 series is the first independent non-MD-Anderson multivariable Cox model retrieved here: SO associated with DFS and OS, LVSI with OS only, but there was no central review, 10 deaths, and wide OS CIs that the authors treat as exploratory (PMID 42794871). Yuan’s LVSI HR 11.95 is n=49 (PMID 31139558). A pooled multi-institutional dataset with path review, SO, LVSI, invasion, age and stage in one model is still the missing study. → staging-and-prognostic-factors, sarcomatous-overgrowth
OQ-2. Does adjuvant chemotherapy change outcome in stage I sarcomatous overgrowth?¶
Carroll’s numerical PFS/OS benefit is not significant (p=0.28 / 0.18) and is the entire comparative signal (PMID 25449308). ESGO allows considering chemo after morcellated high-grade/SO and in resected stage II–IV; French GSF says the role is not established (PMID 39322612, 37202293). A GCIG/EURACAN registry-randomised or carefully matched comparison is the feasible design; a classical phase 3 is not. → adjuvant-and-systemic-therapy, guidelines
OQ-3. Can preoperative diagnosis be made reliably enough to change the morcellation decision for this histology?¶
Occult uterine sarcoma 1/278 is a class estimate (PMID 26646120); adenosarcoma often presents as a polyp (PMID 2156771). MRI features have been catalogued (Morikawa 2025, PMID 39729099) and compared with polyps (Harada 2025, PMID 40210681) without a locked diagnostic threshold or a hysterectomy-gold-standard accuracy study. An adenosarcoma-specific morcellation outcome does not exist (one case-report PubMed hit, PMID 27769260). → clinical-presentation-and-imaging, red-flags-and-safety-concerns
OQ-4. Do any molecular findings support a diagnostic or prognostic test yet?¶
Howitt: no diagnostic tool (PMID 25231023). BAP1 loss ~25% and relatively specific among 196 gyn mesenchymal tumours (PMID 36138078) is the closest marker and is untested in the atypical-polyp differential. TP53/p53 identifies high-grade disease (PMID 28834809). DICER1 is useful only when negative, to make ERMS unlikely (PMID 31900434). ATRX as a transformation marker is n=3 (PMID 36138078). Sharpened 2026-10-17: two case reports of SMARCA4-deficient tumours with adenosarcoma-like architecture (PMIDs: 38113031, 42862247) raise whether SMARCA4/BRG1 IHC belongs in the work-up of high-grade adenosarcoma with rhabdoid stroma in young patients; the frequency of SMARCA4 loss in conventional adenosarcoma is unmeasured (n=2 retained, PMID 34271067). → molecular-and-genomic-features
OQ-5. Who, if anyone, should have a lymphadenectomy?¶
Nodal metastasis 2.9–3.1% (PMID 28109626, 27771166). When present, CSS aHR 2.34, comparable to ESS/LMS (PMID 28109626). SO, size and deep invasion associate with nodal risk in a 230-case literature review. Whether dissection changes outcome is untested. A preoperative/intraoperative rule could be specified and validated. → surgical-management
OQ-6. What would an adequately powered interventional study even look like?¶
263 PubMed records for uterine adenosarcoma (2026-10-31); largest institutional series 165; largest dataset 2,205 NCDB without path review. SEER 22 now supplies a hysterectomy-corrected adenosarcoma incidence of 0.1 per 100,000 without central review (Dave 2026, PMID 42951232). FUCHSia analysed 15 across several histologies; its full text identifies one adenosarcoma partial responder, resolving the enrolment/response question but leaving benefit unquantified beyond n=1 (Trozzi 2024, PMID 38724237; NCT03926936). EORTC 62113 EU registry results report 59 randomised participants versus 58 in ClinicalTrials.gov, without histology-specific baseline counts (NCT01979393; EudraCT results, accessed 2026-10-24). Elacestrant NCT07467772 estimates 30 across four uterine-sarcoma subtypes. The design that matches the disease is a rare-tumour consortium basket with pre-specified adenosarcoma reporting, or a registry-randomised adjuvant question in stage I SO — not a stand-alone phase 3. → clinical-trials-landscape
Tier 2 — by theme¶
Pathology and diagnosis¶
- OQ-7. Surveillance: does lung imaging at defined intervals in SO-positive disease change outcome, or only the time metastasis is labelled? Tate: 75% of distant recurrences included lung, in a literature-assembled series (PMID 29441675). Clement: late local recurrence (PMID 2156771). No schedule has been tested. → recurrence-and-surveillance
- OQ-8. Does BSO change outcome in premenopausal stage I disease without SO? Nasioudis could not conclude for this histology (PMID 28541635); Li n=6 with no recurrences (PMID 35005157). Hormone-receptor biology makes this different from LMS. → surgical-management, patient-experience-and-advocacy
- OQ-9. Can a mitotic/cellularity/copy-number rule separate adenosarcoma from adenofibroma and from atypical polyps with a known error rate? Gallardo’s two fatal “adenofibromas” (PMID 18941402) versus Chapel’s 150-month benign follow-up of atypical polyps that share copy-number events (PMID 34675347). → pathology-and-diagnosis
- OQ-10. Is 25% the right SO cut? Historical (PMID 2535774). SO as a continuous volume fraction against survival has not been modelled. High-grade stroma without SO is a 9-case argument (PMID 28834809). → sarcomatous-overgrowth
Registries and measurement¶
- OQ-11. How much do registry 5-year survival figures (Arend stage I 79%; Dave all-stage relative survival 78.3%) mislead once SO is restored? Dave 2026 now reports hysterectomy-corrected adenosarcoma incidence 0.1 per 100,000 in SEER 22 (1,884 cases, 12.7% of mesenchymal uterine sarcomas) without central review or an SO-stratified survival table in the retrieved abstract (PMID 42951232). Opposite biases remain: registries miss SO; literature-assembled series (Tate 34% death) enrich for published catastrophe (PMID 20688363, 29441675, 33680946). A pathology-linked registry is still the design. → epidemiology-and-burden
Experience¶
- OQ-13. Do adenosarcoma HRQoL priorities differ from LMS enough to need a histology-specific module? den Hollander cannot answer (n=13 mixed; PMID 35443673). Items on sarcoma knowledge, diagnostic shock and menopause are already missing from EORTC tools. → patient-experience-and-advocacy
Closed¶
OQ-12. Did FUCHSia enrol any adenosarcomas, and did they respond? — answered 2026-10-24¶
Yes: one low-grade adenosarcoma without SO, with a partial response (Trozzi 2024, PMID 38724237; NCT03926936). Table 1 and the Efficacy section were checked in the full paper attached to EudraCT, accessed 2026-10-24. This closes the factual enrolment/response question; replication and magnitude of benefit remain open under OQ-6. The registry n=17 versus paper 16 enrolled/15 analysed is unresolved.