0.2% of profiled sarcomas, median age 38, and every number here has a denominator that is not the population
also called KMT2A sarcoma prevalence · KMT2A-rearranged sarcoma incidence · KMT2A sarcoma age distribution
About two in every thousand sarcomas sent for gene sequencing carry a KMT2A rearrangement, and the typical patient is around forty rather than a baby. Both numbers come from tumours that doctors chose to sequence, not from a whole population, so they are floors rather than true rates.
The same thing, for a scientist
There is no population-based incidence figure for KMT2A-rearranged sarcoma. The best available prevalence is 34 of 14,680 sarcomas (0.2%) in a commercial comprehensive genomic profiling database, of which 16 (47%) were YAP1-partnered and 2 VIM-partnered; the remaining 16 had unique non-recurrent partners and are of unclear significance, so the entity-level prevalence is closer to 18/14,680 (0.12%) than to 0.2%. Age: median 38 (22 cases, largest series), median 40 with range 20-66 (pan-sarcoma YAP1 arm), range 10-86 with median 45 (nine-case series), range 11-91 (eight-case series). Sex ratio approximately 1.4:1 male:female. Every denominator is a testing cohort and every series is drawn from consultation or commercial referral, so all figures are subject to ascertainment bias in the same direction - toward aggressive, diagnostically difficult disease. Two further reasons the prevalence is a floor: break-apart FISH is frequently negative in sequencing-confirmed cases, and KMT2A is often not covered on solid-tumour panels. Literature size, measured 2026-09-21: 55 records for KMT2A[tiab] AND sarcoma[tiab], 61 for MLL[tiab] AND sarcoma[tiab], 16 for KMT2A[tiab] AND soft tissue[tiab], 14 for the fusion-name strings.
Every number and how it was measured
| figure | value | denominator | what it is NOT |
|---|---|---|---|
| KMT2A fusion prevalence | 34/14,680 = 0.2% | sarcomas submitted for commercial CGP | not incidence; not population-based; the database is enriched for advanced disease by the authors' own statement 1 |
| entity-level prevalence | 18/14,680 = 0.12% (YAP1 16 + VIM 2) | same | computed here by excluding the 16 non-recurrent-partner cases the authors call of unclear significance 1 |
| YAP1 share of KMT2A-rearranged sarcomas | 47% (16/34) | KMT2A-rearranged sarcomas | not a share of sarcomas 1 |
| median age | 38 y | 22 YAP1::KMT2A cases | not a range; the range across all series is 10-91 2 |
| age range, YAP1 arm of CGP cohort | 20-66 y, median 40 | 16 cases | 1 |
| age range, consultation series | 10-86 y, mean and median 45 | 9 cases | includes a 10-year-old with a dural primary 3 |
| age range, second series | 11-91 y | 8 cases | 7 YAP1::KMT2A + 1 PRRX1::KMT2D 4 |
| M:F | 1.4:1 | 22 cases | 2 |
| local recurrence | 3/14 = 21% | patients with follow-up in the 22-case series | median follow-up 25 months, range 1-170 2 |
| metastasis | 7/14 = 50% | same | same short and uneven window 2 |
| non-recurrent-partner group | median age 57, TP53 mutant 44% (7/16) | 16 cases | the group that inflates the 0.2% 1 |
Literature size, so the next beat knows what it is walking into
Measured by PubMed esearch on 2026-09-21, each query run with one field tag per term and the querytranslation read back 5:
| query | records |
|---|---|
KMT2A[tiab] AND sarcoma[tiab] | 55 |
MLL[tiab] AND sarcoma[tiab] | 61 |
KMT2A[tiab] AND "soft tissue"[tiab] | 16 |
YAP1[tiab] AND KMT2A[tiab] | 17 |
fusion-name strings ("YAP1-KMT2A", "KMT2A-YAP1", "VIM-KMT2A") | 14 |
"sclerosing epithelioid fibrosarcoma"[tiab] | 222 |
"sclerosing epithelioid fibrosarcoma"[tiab] AND KMT2A[tiab] | 14 |
The entire on-topic literature is about a dozen primary reports. Note the MLL AND sarcoma count is the trap: most of its 61 records are myeloid sarcoma (extramedullary leukaemia), which is a leukaemic deposit and not a soft-tissue sarcoma at all. Any census built on that query without title-checking would merge two unrelated diseases.
What this implies for the index fusion, by contrast
Infant KMT2A::AFF1 B-ALL has a population incidence, national registries, cooperative-group trials and a prenatal-origin literature built on twin concordance and neonatal blood spots. This entity has none of those: no registry, no incidence, no trial, no cell line, and a total published experience in the low hundreds of cases at most. The asymmetry is not because the sarcoma is rarer in absolute terms - infant KMT2A-r ALL is itself rare - but because the sarcoma has no cooperative group, no screening programme and a gene that its own diagnostic panels frequently do not include.
| metric | prevalence |
|---|---|
| value | 0.2 |
| unit | percent of sarcomas submitted for comprehensive genomic profiling |
| numerator | 34 |
| denominator | 14680 |
| entity level value | 0.12 |
| entity level note | computed here as (16 YAP1 + 2 VIM)/14,680; the other 16 cases had unique non-recurrent partners the source authors call of unclear significance |
| population | sarcomas interrogated by targeted next-generation DNA and/or RNA sequencing in a commercial comprehensive genomic profiling database |
| region | not stated (commercial US-based profiling laboratory) |
| year | 2020 |
| source | Massoth et al., Modern Pathology 2020 (PMID 32461620) |
| age median years | 38 |
| sex ratio m f | 1.4 |
| recurrence PCT | 21 |
| metastasis PCT | 50 |
| followup median months | 25 |
| is population based | false |
| retrieval hazard | MLL[tiab] AND sarcoma[tiab] is dominated by MYELOID sarcoma (extramedullary leukaemia), not soft tissue sarcoma; the two must be separated by title check |
| sex ratio note | DISCORDANT between the two largest series and should not be quoted as a single figure: the 22-case series reports male:female 1.4:1, while the 16-case YAP1 arm of the pan-sarcoma screen reports 56% FEMALE. At these n both are indistinguishable from 1:1 and neither supports a sex bias in either direction. |
age range years
[ 10, 91 ]
followup range months
[ 1, 170 ]
why a floor
[ "break-apart FISH for KMT2A and YAP1 is frequently negative in sequencing-confirmed cases", "KMT2A is often not covered on solid-tumour sequencing panels", "the source database is enriched for advanced disease by the authors' own statement" ]
literature size 2026 09 21 — the numbers behind the chart
{
"KMT2A[tiab] AND sarcoma[tiab]": 55,
"MLL[tiab] AND sarcoma[tiab]": 61,
"KMT2A[tiab] AND \"soft tissue\"[tiab]": 16,
"YAP1[tiab] AND KMT2A[tiab]": 17,
"fusion name strings": 14,
"\"sclerosing epithelioid fibrosarcoma\"[tiab]": 222,
"\"sclerosing epithelioid fibrosarcoma\"[tiab] AND KMT2A[tiab]": 14
}- A 0.2% prevalence in a profiling cohort is the number most likely to be quoted as an incidence, and it is not one. Any addressable-population estimate for a drug programme built on it would be wrong in an unknown direction.
- The 50% metastasis figure comes from 14 patients over a median 25 months and would look very different at 5 years. Using it as a benchmark for single-arm activity would set an efficacy bar that the natural history could clear or fail by chance.
- What is the population incidence? A sarcoma registry with universal RNA sequencing of MUC4-negative, EWSR1/FUS-negative SEF-like tumours would give the first real denominator, and nothing smaller will.
- How many MUC4-negative SEF cases in existing archives are actually this entity? The founding series predicts the answer is 'more than currently counted' and the experiment is retrospective RNA sequencing of stored blocks - no new patients needed.
- Is the paediatric end of the age range real? The youngest reported case (10 years, dural) was identified by the assay this same literature shows to be unreliable.
- Does the entity have a sex bias? 1.4:1 in 22 cases is indistinguishable from 1:1 at that n, and reporting it as a bias would be a statement about the sample.
- The two largest series disagree on the direction of the sex ratio (1.4:1 male versus 56% female). At n=22 and n=16 that is what 1:1 looks like, and this graph should carry no sex-bias claim until a larger series exists.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In a comprehensive genomic profiling database of 14,680 human sarcomas, 34 KMT2A-fusion-positive cases (0.2%) were identified; YAP1 was the partner in 16 (47%) affecting patients aged 20-66 (median 40), VIM in 2 (men aged 43 and 46), and the remaining 16 had unique non-recurrent partners with a median age of 57, diverse histologies and pathogenic TP53 mutation in seven (44%).
Pan-sarcoma genomic analysis of KMT2A rearrangements reveals distinct subtypes defined by YAP1-KMT2A-YAP1 and VIM-KMT2A fusions. · Massoth LR, Hung YP, Nardi V, et al. · Mod Pathol 2020What this evidence cannot show
The denominator is sarcomas submitted for commercial profiling, which the authors state is enriched for aggressive and advanced disease; it is not an incident population. KMT2A is often not covered on solid-tumour panels and FISH is insensitive here, so 0.2% is a floor of unknown depth. The 34 comprises one separately identified index patient plus 33 database cases (the methods state 33 after excluding myeloid/monocytic sarcoma).
Evidence that argues against this (1)
- The authors of a companion series state explicitly that because their sequencing-confirmed cases were FISH-negative, the prevalence of this fusion among MUC4-negative sclerosing epithelioid fibrosarcoma is expected to be higher than measured. Any prevalence figure produced by a panel that frequently misses the lesion understates it. Kao YC et al. 2020
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In the largest series of human YAP1::KMT2A-rearranged sarcomas (22 cases), median age was 38 years with a male:female ratio of 1.4:1, and among 14 patients with follow-up at a median of 25 months (range 1-170) local recurrence occurred in three (21%) and metastasis in seven (50%).
YAP1::KMT2A-rearranged sarcomas harbor a unique methylation profile and are distinct from sclerosing epithelioid fibrosarcoma and low-grade fibromyxoid sarcoma. · Warmke LM, Ameline B, Fritchie KJ, et al. · Virchows Arch 2025What this evidence cannot show
Retrospective multi-institutional collection; follow-up available in 14 of 22, at a median of only 25 months with a range spanning 1 to 170, so the 21% and 50% are pooled over wildly unequal exposure times and are not actuarial estimates. Several cases had initially false-negative FISH and panel results, so the series consists of cases that survived a diagnostic gauntlet.
Evidence that argues against this (1)
- Both single-case reports of non-YAP1, non-VIM arrangements (CBX6::KMT2A::PYGO1 and LDB1::KMT2A) were disease-free after surgery alone at 32 and 5 months. Either behaviour tracks the partner, or the aggressive figures reflect series assembled from consultation and commercial profiling. With n=1 per rare partner neither can be separated. Wang Z et al. 2025
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In nine human MUC4-negative sclerosing epithelioid fibrosarcomas with YAP1 and/or KMT2A rearrangements, patients ranged from 10 to 86 years (mean and median 45), four female and five male, including a 12 cm frontal dural tumour in a 10-year-old girl and a 22-year-old man with a KMT2A(exon 6)-YAP1(exon 9) fusion.
Recurrent YAP1 and KMT2A Gene Rearrangements in a Subset of MUC4-negative Sclerosing Epithelioid Fibrosarcoma. · Kao YC, Lee JC, Zhang L, et al. · Am J Surg Pathol 2020What this evidence cannot show
Nine cases from consultation practice, so the age distribution is not representative of the entity. Only some cases had sequencing-level confirmation; the rest were identified by FISH, which this same paper shows to be insensitive - so the age range may include cases that are not this entity. Read from the PMC full text via efetch db=pmc.
Evidence that argues against this (1)
- The 10-year-old case was identified by YAP1 FISH rather than by RNA sequencing, and the paper shows FISH to be unreliable in both directions in this entity, so the paediatric end of the age range rests on the weaker assay. Warmke LM et al. 2025
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In eight human soft tissue neoplasms with morphology overlapping low-grade fibromyxoid sarcoma and sclerosing epithelioid fibrosarcoma but lacking MUC4 and EWSR1/FUS-CREB3L fusions, seven showed YAP1-KMT2A fusions and one a PRRX1-KMT2D fusion, in four women and four men aged 11 to 91 years, with tumour sizes of 2.5 to 13 cm and gene expression profiles distinct from LGFMS/SEF.
Recurrent Fusions Between YAP1 and KMT2A in Morphologically Distinct Neoplasms Within the Spectrum of Low-grade Fibromyxoid Sarcoma and Sclerosing Epithelioid Fibrosarcoma. · Puls F, Agaimy A, Flucke U, et al. · Am J Surg Pathol 2020What this evidence cannot show
Eight cases with follow-up in five (11-321 months). Read from the abstract. The single PRRX1-KMT2D case is a different gene (KMT2D, not KMT2A) and is recorded here only because it appears in the same cohort; this graph's KMT2D/MLL2-MLL4 naming hazard applies and the case was not pursued.
Evidence that argues against this (1)
- This series found LGFMS-like histology in three of seven cases whereas two other series saw none, so the morphological spectrum of the YAP1 arm is not agreed even among the founding reports. An age or size range drawn from a cohort whose inclusion morphology is contested inherits that uncertainty. Massoth LR et al. 2020
- E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
Measured here by PubMed esearch on 2026-09-21 with one field tag per term and the querytranslation read back: KMT2A[tiab] AND sarcoma[tiab] returns 55 records, MLL[tiab] AND sarcoma[tiab] returns 61, KMT2A[tiab] AND "soft tissue"[tiab] returns 16, YAP1[tiab] AND KMT2A[tiab] returns 17, and "sclerosing epithelioid fibrosarcoma"[tiab] returns 222 of which 14 also mention KMT2A.
PubMed E-utilities esearch, KMT2A-sarcoma literature size measurements · National Center for Biotechnology Information · PubMed E-utilities 2026What this evidence cannot show
Counts are of titles and abstracts only, on the dates stated; full texts and supplementary figures are not searched by esearch. The MLL AND sarcoma count of 61 is heavily contaminated by MYELOID sarcoma (extramedullary leukaemia), a different disease, and was not title-checked case by case - it is quoted as a retrieval size, not as a count of on-topic papers. The queries use one tag per term precisely because the parenthesised-group form silently drops the tag.
Evidence that argues against this (1)
- A literature size is not a measure of how much is known. Of these dozen-odd primary reports, all are pathology case series; not one is a functional study in a model system, so the bibliographic size overstates the mechanistic knowledge by a wide margin. Teku G et al. 2023
prevalence, age, sex and outcome figures with their denominators
the same gene rearranged four decades earlier in life, with a registry-based incidence this entity has no counterpart for
16 of 34 KMT2A-rearranged sarcomas, median age 40, range 20-66
2 of 34, men aged 43 and 46