Jump to content
Content on WikiMesothelioma is reviewed prior to publication by three named attorneys at Danziger & De Llano LLP. Dr. Luciano Mutti, MD, PhD, co-author of the European mesothelioma treatment guidelines, serves as the site's Medical Reviewer. See our editorial standards.

Mesothelioma Diagnostic Technology

From WikiMesothelioma — Mesothelioma Knowledge Base
Mesothelioma Diagnostic Technology (2025–2026)
Current staging IASLC TNM 9th ed. (Jan 2025) — pleural thickness sum (Psum)
Tissue gold standard BAP1 IHC + MTAP IHC (100% specificity for malignancy)
FDA-approved serum biomarker MESOMARK (SMRP) — monitoring only, not diagnosis
Methylation profiling (emerging) IMPRESS tissue methylation — 100% specificity for PM vs. pleural metastases (research validation study)
AI pathology SpindleMesoNET AUC 0.989 (external test set); MesoNet validated; no FDA-cleared mesothelioma-specific tool
Germline testing BAP1 germline testing — offered to all patients per ASCO 2025 Rec 7.1 (High, Strong); European genetics consensus advises against routine testing in isolated mesothelioma
Histology classification WHO 5th ed. (2021) — epithelioid, biphasic, sarcomatoid; WDPMT reclassified non-malignant
Verified 2026-05-13

Executive Summary

The 2023–2026 period delivered the largest set of mesothelioma diagnostic advances in two decades. Four shifts now define current practice:

The IASLC TNM 9th edition, published January 2025, replaced anatomically defined T1–T4 invasion descriptors with pleural thickness sum (Psum) measured at three axial CT levels — a continuous, reproducible variable that stratifies median overall survival from 49.8 months at Psum ≤12 mm to 21.1 months at Psum >31 mm.

In pathology, MTAP immunohistochemistry has emerged as a practical surrogate for CDKN2A homozygous deletion, supplementing BAP1 IHC as the two-marker rule-in panel for malignant mesothelial proliferation. MTAP loss is a surrogate for CDKN2A homozygous deletion, and BAP1 loss, CDKN2A homozygous deletion and MTAP loss are highly specific markers of malignancy in a mesothelial lesion.[1] On effusion cell blocks the evidence is weaker: one study showed a specificity of 86% for BAP1 IHC on cytology, and caution is warranted when cytology is the basis of a primary diagnosis.[1]

The WHO 5th Edition Classification of Thoracic Tumours (2021) reclassified well-differentiated papillary mesothelial tumor (WDPMT) as a non-malignant entity and formally recognized mesothelioma in situ as a diagnostic category requiring BAP1 loss and/or CDKN2A deletion.

The IMPRESS tissue methylation assay demonstrated 100% specificity for distinguishing pleural mesothelioma from pleural metastases.[2] ASCO 2025 Recommendation 7.1 states that all patients with mesothelioma should be offered germline testing (evidence quality High; strength of recommendation Strong).[3] The European BAP1-TPDS consensus is more restrictive, advising that testing "should not routinely be offered" in isolated mesothelioma (95% consensus, Moderate /weak).[4] No new serum biomarker has received FDA approval since MESOMARK in 2007.

At a Glance

  • Staging: IASLC TNM 9th edition (January 2025) uses pleural thickness sum on CT, replacing 8th edition anatomical T-categories.
  • Tissue diagnosis: BAP1 IHC loss + MTAP IHC loss = 100% specificity for malignant mesothelial proliferation.
  • Imaging first-line: Multi-detector CT (sensitivity 68%, specificity 78%); MRI superior for chest-wall and diaphragm invasion; PET-CT sensitivity ~95% for malignant pleural disease but limited (~30% nodal upstaging at surgery).
  • Liquid biopsy: MESOMARK (SMRP) is the only FDA-approved serum biomarker (monitoring, not diagnosis); cfMeDIP-seq, IMPRESS methylation, fibulin-3, and ctDNA are investigational.
  • AI pathology: SpindleMesoNET (AUC 0.989 on its external test set) performed comparably to expert pathologists for sarcomatoid vs. benign spindle classification; MesoNet (Nature Medicine 2019) predicts outcome from whole-slide images. No mesothelioma-specific AI tool has FDA clearance.
  • Germline testing: ASCO's 2025 guideline update, Recommendation 7.1, states that all patients with mesothelioma should be offered germline testing (evidence quality High; strength of recommendation Strong); European genetics guidance advises against offering it routinely in isolated mesothelioma. Where testing is done it guides surveillance and cascade screening for blood relatives, and does not establish what caused the patient's own disease.
  • Differential diagnosis: BAP1 + MTAP (or CDKN2A FISH) is the core panel to distinguish malignant mesothelioma from reactive mesothelial proliferation; solitary fibrous tumor identified by STAT6 / NAB2-STAT6 fusion.
  • Commercial BAP1 assay: On October 28, 2026, Roche announced the global launch of the VENTANA BAP1 (SP527) Rabbit Monoclonal Primary Antibody, an automated IHC assay registered with the FDA as a Class I device, a class exempt from premarket notification; FDA rules state that registration does not denote approval.[5][6][7][8]
  • Subtype: Epithelioid (~55–60%), biphasic (~20–25%), sarcomatoid (~10–20%); sarcomatoid carries the worst prognosis and per ASCO 2025 contraindicates maximal surgical cytoreduction.

Key Facts

Diagnostic Tool FDA Status Best Performance Primary Use Guideline Status (2026)
MESOMARK (SMRP, serum) Approved 2007 Pooled sensitivity 47% / specificity 96% (study ranges 19–68% / 88–100%)[9] Monitoring epithelioid or biphasic MPM ASCO 2025: non-tissue biomarkers not recommended to predict outcome or monitor tumor response (Rec. 6.8)[10]
Fibulin-3 (plasma) Not approved See Fibulin-3 section below Investigational ASCO 2025: non-tissue biomarkers not recommended to predict outcome or monitor tumor response (Rec. 6.8)[10]
IMPRESS (tissue methylation) Research validation study Specificity 100% for PM vs. pleural metastases Distinguishing PM from other pleural pathologies ⚠️ Emerging[2]
BAP1 IHC (tissue or cytology) Clinical use Virtually 100% specific for malignancy in a mesothelial proliferation; sensitivity 50–65% (pleural)[1] Rule-in malignancy IMIG / WHO endorsed
MTAP IHC (CDKN2A surrogate) Clinical use ~95% concordance with CDKN2A FISH Rule-in malignancy ⚠️ Emerging in IMIG 2024 practice updates
CDKN2A FISH Clinical use Virtually 100% specific for malignancy in a mesothelial proliferation; sensitivity 50–65% (pleural), 80–100% (sarcomatoid)[1] Rule-in malignancy IMIG / WHO endorsed
MesoNet AI (WSI) Research only Improves OS prediction beyond pathologist annotation Outcome prediction Research[11]
SpindleMesoNET (CNN) Research only AUC 0.932 / 0.925 / 0.989 (cross-validation / referral / external test sets); referral-set accuracy 92.5% (vs. 91.7% expert average) Sarcomatoid vs. benign spindle Research[12]
PET-CT (¹⁸F-FDG) Standard of care Sensitivity ~95% for malignant pleural disease; ~30% nodal upstaging at surgery Staging NCCN / ASCO endorsed
TNM 9th edition (Psum) In use Jan 2025 Median OS 49.8 mo (T1) → 21.1 mo (T3) Pleural mesothelioma staging IASLC / UICC endorsed

What Are the Latest Imaging Advances for Mesothelioma in 2025–2026?

Multi-detector CT (First-Line)

CT scanning has a sensitivity of approximately 68% and specificity of approximately 78% for detecting pleural malignancy, and cannot reliably differentiate diffuse pleural mesothelioma from metastatic pleural disease.[13] Multi-detector CT (MDCT) remains the first-line modality for suspected pleural mesothelioma. Classic findings include unilateral pleural effusion, nodular or circumferential pleural thickening greater than 1 cm, a pleural rind encasing the lung, and involvement of the mediastinal or diaphragmatic pleura.[13] CT cannot reliably assess chest-wall invasion (the T3/T4 boundary) or transmural pericardial involvement — limitations that drive the use of MRI for surgical candidates.

18F-FDG PET-CT

PET-CT distinguishes malignant from benign pleural lesions with a sensitivity of approximately 95% and specificity of approximately 82% for malignant pleural disease.[13] PET-CT sensitivity for mediastinal lymph node staging is limited: pathological upstaging of the N-category occurs in approximately 30% of cases at surgery, making PET-CT insufficient as a standalone staging tool for surgical candidates. False positives include tuberculous pleurisy, prior talc pleurodesis (with SUV reaching 3–5 at pleurodesis sites for months), empyema, and asbestos-related benign pleuritis.

MRI for Chest-Wall and Diaphragm Assessment

MRI is superior to CT for evaluating chest-wall invasion, diaphragmatic muscle invasion, and pericardial involvement — all critical T3/T4 distinctions in the new staging system. Diffusion-weighted MRI (DWI) identifies areas of restricted diffusion corresponding to viable tumor, and MRI is the standard modality of choice when surgical candidacy depends on resolution of soft-tissue invasion that CT cannot reliably assess. Hyperpolarized MRI using ¹³C-pyruvate has shown proof-of-concept in early-phase UK lung and pleural malignancy trials but no mesothelioma-specific hyperpolarized MRI diagnostic data have been published in peer-reviewed literature as of May 2026 — it remains investigational.

TNM 9th Edition: Pleural Thickness Measurement (Major 2025 Change)

The IASLC 9th edition TNM for pleural mesothelioma, published January 2025, introduced pleural thickness sum (Psum) measured at three axial CT levels — apex, hilum, and base — as the primary T-descriptor, replacing the 8th edition's anatomically defined T1–T4 invasion-based categories. The change was driven by data showing the 8th edition T-categories had poor inter-observer reproducibility and that pleural thickness was a more continuous, objectively measurable prognostic variable. The new descriptors stratify median overall survival sharply:

  • T1 (Psum ≤12 mm) — median overall survival 49.8 months
  • T2 (Psum 12–30 mm) — median overall survival 27.5 months
  • T3 (Psum >31 mm) — median overall survival 21.1 months

The N-descriptor was largely retained (ipsilateral intrathoracic nodes = N1; contralateral or supraclavicular = N2). The 8th edition stage groupings remain in widespread use during the 2025–2026 transition.

Which Liquid Biopsy Tools Are Available for Mesothelioma?

The MESOMARK assay (Fujirebio Diagnostics) is the only FDA-approved serum biomarker for mesothelioma as of May 2026. FDA approval was granted January 24, 2007 for monitoring disease response or progression in epithelioid or biphasic pleural mesothelioma — not for primary diagnosis or screening.

Diagnostic performance across major studies has been heterogeneous:

  • Robinson et al., The Lancet, 2003 — raised serum SMR in 37 of 44 patients with mesothelioma (84%), against 3 of 160 (2%) with other inflammatory or malignant lung and pleural diseases[14]
  • Hollevoet et al. individual patient data meta-analysis of 16 diagnostic studies (4,491 individuals) — pooled sensitivity 47% (range 19–68%) at specificity 96% (range 88–100%)[9]
  • At a predefined 99% specificity (screening context), sensitivity drops to 23%

The SOMAscan SOMAmer-based proteomics platform differentiated pleural mesothelioma from asbestos-exposed controls in the DIAPHRAGM study with sensitivity 75% and specificity 88.2%, but was not useful for distinguishing mesothelioma from non-mesothelioma patients presenting with suspected pleural malignancy.

Fibulin-3 (EFEMP1) — Not FDA Approved

Fibulin-3 was initially reported with extraordinary diagnostic performance in the 2012 Pass et al. study (AUC 0.99, sensitivity 95%, specificity 95% for mesothelioma vs. asbestos-exposed controls).[15] Subsequent validation has not replicated the original results. In the multicenter DIAPHRAGM study (638 patients with suspected pleural malignancy, 112 asbestos-exposed controls), fibulin-3 demonstrated only 7.4%–11.9% sensitivity at 95% specificity — significantly inferior to mesothelin. A 2025 prospective study by Jiang et al. (Radiology and Oncology, 90 thoracoscopic-biopsy patients) found plasma fibulin-3 at a 12.31 ng/mL cutoff offered 100% specificity but only 39.39% sensitivity for distinguishing mesothelioma from asbestos-related benign pleural disease (AUC 0.78). A composite panel of serum fibulin-3 + mesothelin + HMGB1 achieved sensitivity 96% and specificity 93% in one retrospective study — substantially outperforming any single marker, but requires prospective validation.

HMGB1 (High Mobility Group Box 1) — Investigational

HMGB1, particularly its hyperacetylated isoform, has shown promise as a serum biomarker for detecting asbestos exposure and early mesothelioma via the RAGE receptor/Beclin-1 autophagy pathway. HMGB1 is not FDA-approved for any diagnostic indication in mesothelioma as of May 2026. Large-scale prospective validation is pending.

ctDNA and Methylation-Based Liquid Biopsy (Emerging)

Circulating tumor DNA (ctDNA) profiling faces unique obstacles in mesothelioma due to the absence of recurrent gain-of-function driver mutations — the disease is driven predominantly by loss of tumor suppressor genes (BAP1, NF2, CDKN2A), making conventional mutation-based ctDNA panels poorly suited for detection. The most promising approach is epigenetic methylation-based ctDNA analysis. One such method is cfMeDIP-seq (plasma cell-free methylated DNA immunoprecipitation sequencing).

The IMPRESS tissue assay (bisulfite-free methylation detection, 744 hypermethylated CpG sites), published in Molecular Oncology in 2026, achieved:

  • Sensitivity 89.2% and specificity 93.5% for differentiating tumoral from non-tumoral pleura
  • Sensitivity 85.2% and specificity 100% for distinguishing pleural mesothelioma from pleural metastases[2]

A 2025 Nature Medicine Phase 2 study (Johns Hopkins) demonstrated that tumor-informed ctDNA detection can monitor neoadjuvant immunotherapy response in resectable diffuse pleural mesothelioma. Using concurrent tumor and cell-free DNA whole-genome sequencing with machine learning, persistent ctDNA at cycle 3 was associated with shorter PFS (log-rank p=0.0059 pre-surgery), and detectable ctDNA predicted failure to achieve complete surgical resection (Fisher's exact p=0.00013). No liquid biopsy test has received FDA approval specifically for mesothelioma as of May 2026.

How Are Tissue Biomarkers Used to Diagnose Mesothelioma?

WHO 5th Edition Classification (2021) — Key Updates

The WHO Classification of Thoracic Tumours, 5th Edition (2021, IARC Press) introduced critical changes affecting diagnosis:

  • Well-differentiated papillary mesothelial tumor (WDPMT) was reclassified as a non-malignant entity (borderline / uncertain malignant potential), distinct from diffuse pleural mesothelioma. WDPMT is associated with TRAF7 and CDC42 mutations (rather than BAP1/CDKN2A alterations), confirming its non-malignant biology — unnecessary chemotherapy should be avoided in confirmed WDPMT cases.
  • Mesothelioma in situ was formally recognized as a diagnostic category requiring BAP1 loss and/or CDKN2A deletion for confirmation.
  • Localized malignant mesothelioma was retained as a distinct entity, separable from diffuse disease.

Standard IHC Panel: Positive Mesothelial Markers

Per IMIG / WHO consensus, the diagnostic IHC approach uses at least two mesothelial-positive markers AND two epithelial-negative markers, each with >80% sensitivity and specificity, selected based on morphology and clinical context.

Marker Overall Sensitivity Epithelioid Sarcomatoid Notes
Calretinin ~87.6% 93.2% 16.2% Nuclear/cytoplasmic; best single marker
WT1 ~77% High Very low Nuclear; strong for epithelioid
D2-40 (podoplanin) ~75% Moderate–high Lower Membranous
CK5/6 Variable ~70% ~10% Less subtype-specific
Mesothelin ~70% High Very low Also expressed in lung adenocarcinoma

Standard Negative (Adenocarcinoma) Markers

  • Claudin-4 — highly specific for carcinoma, absent in mesothelioma
  • BerEP4 / MOC-31 — positive in adenocarcinoma, negative in mesothelioma
  • TTF-1 / Napsin A — lung adenocarcinoma markers, negative in mesothelioma
  • CEA — positive in GI-origin adenocarcinoma, negative in mesothelioma

BAP1 IHC — The Key Malignancy Discriminator

Loss of nuclear BAP1 staining by IHC has virtually complete specificity for malignancy in the mesothelial lineage — it is the single most important biomarker for distinguishing malignant from benign mesothelial proliferations:

  • Pooled sensitivity: 56% (95% CI 0.50–0.62)
  • Pooled specificity: 100% (95% CI 0.95–1.00)

The preponderance of evidence indicates that BAP1 IHC can be accurately applied to cell block specimens from effusions, although one study showed a specificity of 86%, so caution is warranted in interpreting BAP1 IHC on cytology specimens, especially for a primary diagnosis.[1] Loss of BAP1 in tumor tissue is a tumor finding used diagnostically, distinct from an inherited variant. Germline BAP1 pathogenic variants are uncommon: a European clinical-genetics consensus records their prevalence in individuals with isolated mesothelioma as low (less than 3%).[4] Their diagnostic relevance is that they flag families for surveillance; they are not a measure of how much of the disease burden is inherited rather than caused by asbestos.[16]

Commercial assay: VENTANA BAP1 (SP527)

On October 28, 2026, Roche announced the global launch of the VENTANA BAP1 (SP527) Rabbit Monoclonal Primary Antibody, a ready-to-use automated IHC assay.[5] Roche says the assay detects nuclear loss of BAP1 expression, a tumor suppressor marker on chromosome 3p21, and that "IHC loss of nuclear BAP1 is virtually 100% specific for malignancy in mesothelial proliferations".[5] Roche positions it alongside lineage markers such as WT1 and calretinin and its VENTANA MTAP (SP530) and Claudin-4 (EP417) tests.[5] Roche reports FDA registration as a Class I in vitro diagnostic and European IVDR approval as a Class C device.[5] The FDA listing (product code NJT, regulation 21 CFR 864.1860, class 1) names Roche Diagnostics GmbH and Ventana Medical Systems.[6] Class I IHC reagents under that regulation are exempt from FDA premarket notification.[7] FDA regulations state that registration of a device establishment "does not in any way denote approval of the establishment or its products."[8]

MTAP IHC as Surrogate for CDKN2A Deletion

MTAP (methylthioadenosine phosphorylase) IHC has been validated as a practical surrogate for CDKN2A homozygous deletion (9p21 locus). Because MTAP is co-located with CDKN2A on chromosome 9p21, homozygous CDKN2A deletion almost invariably results in MTAP protein loss detectable by IHC:

  • MTAP IHC loss sensitivity for CDKN2A homozygous deletion: ~95%
  • MTAP IHC loss specificity: ~95%

MTAP IHC is simpler, faster, and less expensive than CDKN2A FISH. The combination of BAP1 IHC loss + MTAP IHC loss identifies a large proportion of malignant mesothelial proliferations with 100% specificity — replacing the need for CDKN2A FISH in most centers. FISH is now reserved for cases where IHC results are equivocal.

What Molecular and Genomic Tests Are Used for Mesothelioma?

Targeted NGS Panels (2024–2026)

Clinical-grade NGS panels available for mesothelioma profiling:

Panel Key Alterations Detected Mesothelioma Relevance
FoundationOne CDx (Foundation Medicine) BAP1, NF2, CDKN2A, LATS1/2, TP53, TMB, MSI FDA-approved CDx; detects actionable alterations
MSK-IMPACT (Memorial Sloan Kettering) 505-gene panel — BAP1, NF2, CDKN2A, SETD2 Research / clinical use at MSK
Tempus xT 648-gene panel including all MPM tumor suppressors RNA-fusion detection included
Caris Molecular Intelligence Comprehensive — IHC + NGS + FISH Multi-platform profiling

At the ESMO Congress 2026, Memorial Sloan Kettering researchers reported MSK-IMPACT sequencing of tumors from 68 patients with metastatic mesothelioma, 20 of whom had both the original tumor and a metastasis sequenced. Some alterations appeared only in the metastasis, among them TERT and FANCA (3 patients each), but no difference between the grouped primary and metastatic samples was statistically significant, and the authors describe their comparisons with a public tumor dataset as trends that need validation. This is a conference abstract, not a peer-reviewed paper.[17]

The ASCO 2025 Mesothelioma Guideline (Kindler et al.) developed recommendations covering germline testing in pleural mesothelioma.[3] The purpose of the recommendation is forward-looking: a positive result identifies blood relatives who may benefit from surveillance, and can affect treatment selection. It is not a finding about what caused the tested patient's mesothelioma.

BAP1 Germline Testing and BAP1-TPDS

Germline BAP1 pathogenic variants are associated with BAP1 Tumor Predisposition Syndrome (BAP1-TPDS), which is inherited in an autosomal dominant manner.[18] According to GeneReviews, the syndrome is associated with an increased risk of a skin lesion called a BAP1-inactivated melanocytic tumor (BIMT, formerly called an atypical Spitz tumor) and, in descending order of frequency, uveal melanoma, malignant mesothelioma, cutaneous melanoma, renal cell carcinoma (usually of clear cell morphology), basal cell carcinoma, meningioma and cholangiocarcinoma.[18] Identifying the syndrome matters clinically because it tells a family which surveillance is worth doing.

A germline BAP1 variant does not displace asbestos as the cause of a carrier's mesothelioma, and a positive germline result should not be read that way.[16] Asbestos remains the established cause of mesothelioma, including in carriers; the practical significance of the syndrome is surveillance and cascade screening for relatives, not an alternative explanation for the patient's own diagnosis.[16]

European genetics guidance does not agree that every patient should be tested, and the disagreement is between two different kinds of body rather than a misreading by either. The BAP1 tumour predisposition syndrome Delphi consensus records that germline BAP1 genetic testing "should not routinely be offered to all individuals diagnosed with isolated cases of uveal melanoma, mesothelioma, cutaneous melanoma or renal cell carcinoma, but can be considered based on age of onset and other factors ..." — 95% consensus, rated Moderate /weak. The same consensus records at 81% and rated Strong that "the prevalence of germline BAP1 pathogenic variants in individuals with isolated mesothelioma is low (less than 3%)".[4] ASCO is a United States oncology treatment guideline; this is a European clinical-genetics guideline. They address different questions and are cited here as two positions, not one.

Beyond BAP1: Other Inherited Variants (2026 Preprint)

A preprint posted on medRxiv in October 2026, which has not been peer-reviewed, looked beyond BAP1. Researchers led from the University of Turin sequenced the inherited DNA of 127 participants in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort who later developed pleural mesothelioma, using blood drawn before diagnosis. Each was compared with a cancer-free participant matched on age, sex, tumor histotype and asbestos exposure grade.[19] Of the 127 patients, 31 (24%) carried a pathogenic or likely pathogenic variant in a panel of 193 cancer-associated genes, compared with 6% of the 127 matched controls. Protein-truncating variants were more common in patients than in their exposure-matched controls (odds ratio 3.5, 95% CI 1.3–11.2), and carriers lived longer after diagnosis (hazard ratio 0.54, 95% CI 0.32–0.92).[19]

Because patients and controls were matched on asbestos exposure, the study asks why some people with the same exposure develop mesothelioma and others do not. The authors describe two of the genes, BAP1 and POT1, as biologically plausible candidates "for modulating interindividual asbestos risk" and call for validation in larger cohorts.[19] Asbestos remains the established cause of mesothelioma, including in people who carry such variants.[16]

CDKN2A FISH

CDKN2A homozygous deletion detected by fluorescence in situ hybridization (FISH) is virtually 100% specific for malignancy in a mesothelial proliferation; in a published review's pooled ranges, its sensitivity is 50–65% in pleural mesothelioma and 80–100% in sarcomatoid mesothelioma.[1] Combined with BAP1 IHC, the two-test panel reaches a reported sensitivity of 58–93%, with most reported figures in the 85–93% range.[1]

Is AI Used in Mesothelioma Diagnosis?

Regulatory Status

As of May 2026, no AI-assisted pathology tool has received FDA approval specifically for mesothelioma diagnosis. However, several research-stage tools and broader AI platforms are relevant to mesothelioma diagnostic workflows. Paige PanCancer Detect received FDA Breakthrough Device Designation in April 2025 for AI-assisted cancer detection across multiple tissue and organ types — the first multi-site AI pathology designation. While not mesothelioma-specific, the platform could theoretically flag mesothelioma specimens in mixed pathology workflows.

Published AI Models for Mesothelioma

Model Type Task Performance
MesoNet (Courtiol et al., Nature Medicine 2019)[11] Deep-learning CNN Whole-slide outcome prediction without pathologist annotation Improves overall survival prediction; identifies key histological regions
SpindleMesoNET (Naso et al., Modern Pathology 2021)[12] CNN Sarcomatoid mesothelioma vs. benign spindle cell differentiation AUC 0.989 on the external test set; referral-set accuracy 92.5% (vs. 91.7% expert pathologist average)
SVM biomarker models Machine learning Mesothelioma diagnosis from clinical/biomarker data 99.87% accuracy in smaller validation datasets
Multi-model ensembles (DT, SVM, LogR, RF) Ensemble ML Mesothelioma diagnosis Up to 100% accuracy in small validation datasets

FDA-approved AI tools exist for pleural disease detection (e.g., pneumothorax) with sensitivity 84.3–94.6% and specificity 87.9–95.1%, but these are radiology tools — not histopathology tools — and have no mesothelioma-specific validation.

How Is Mesothelioma Distinguished from Other Pleural Diseases?

Key Differential Diagnoses

The principal diagnostic challenge is distinguishing diffuse malignant pleural mesothelioma from:

  1. Pleural metastatic adenocarcinoma (lung, breast, GI) — most common mimic; requires negative epithelial IHC markers (Claudin-4, BerEP4, MOC-31, TTF-1)
  2. Benign reactive mesothelial proliferation — critical distinction requiring BAP1/MTAP IHC + CDKN2A FISH
  3. Fibrous pleurisy — organized inflammatory process; BAP1 / CDKN2A help confirm benign nature
  4. Solitary fibrous tumor — STAT6 nuclear positivity and NAB2-STAT6 fusion gene distinguish from mesothelioma
  5. Well-differentiated papillary mesothelial tumor (WDPMT) — now classified non-malignant (WHO 5th ed.); TRAF7 / CDC42 mutations distinguish from mesothelioma

The Core Benign-vs-Malignant Workflow

Distinguishing reactive mesothelial hyperplasia from early or focal pleural mesothelioma is the most diagnostically challenging scenario in mesothelioma pathology. The BAP1 IHC + MTAP IHC panel (or BAP1 IHC + CDKN2A FISH) is the current standard:

  • If BAP1 loss OR MTAP loss is present → malignant (virtually 100% specificity)
  • If neither loss is present → cannot exclude malignancy; repeat biopsy or clinical correlation required

Pleural fluid cytology is essentially non-diagnostic for sarcomatoid mesothelioma — sarcomatoid cells do not exfoliate into pleural fluid. Sensitivity for sarcomatoid mesothelioma from cytology is approximately 6%. Even for epithelioid mesothelioma, cytology sensitivity is only ~35% overall.[20]

  1. Clinical history and CT → suspect pleural malignancy
  2. Thoracentesis (fluid cytology, with BAP1 IHC on the cell block). In selected patients in whom tissue is not available or cannot be obtained, cytologic diagnoses may be used for treatment decisions[10]; caution is warranted with BAP1 IHC on cytology for a primary diagnosis[1]
  3. If cytology non-diagnostic or sarcomatoid suspected → CT-guided core needle biopsy (16-gauge; diagnostic yield 93%) or medical thoracoscopy / VATS (gold standard; sensitivity 92–95%)
  4. IHC panel on tissue — ≥2 positive mesothelial markers + ≥2 negative epithelial markers
  5. BAP1 IHC + MTAP IHC (or CDKN2A FISH) on all specimens to confirm malignancy
  6. PET-CT for staging (limited nodal accuracy — upstaging ~30% at surgery)
  7. MRI if T3 / T4 chest-wall or diaphragm involvement is suspected
  8. Mediastinoscopy or EBUS-EUS if surgical candidate and PET-CT suggests N1 / N2 disease
  9. Germline BAP1 testing offered to all patients — ASCO 2025 Recommendation 7.1 (evidence quality High; strength of recommendation Strong); not routinely advised in isolated mesothelioma by the European genetics consensus[3][4]
  10. NGS panel for molecular profiling (prognostic, predictive, trial eligibility)

How Does Histologic Subtype Affect Mesothelioma Diagnosis and Treatment?

The Three WHO Subtypes

  • Epithelioid (~55–60%) — polygonal or cuboidal cells with abundant eosinophilic cytoplasm and round nuclei; tubular, trabecular, or solid patterns; best prognosis; most responsive to immunotherapy and chemotherapy; strongest BAP1 IHC positivity
  • Sarcomatoid (~10–20%) — spindle cells resembling sarcoma; subvariants include desmoplastic (pure collagen matrix) and transitional; worst prognosis (median OS 4–6 months); unreliable mesothelial IHC marker expression (calretinin positivity only ~16%)
  • Biphasic (~20–25%) — contains both epithelioid and sarcomatoid components; intermediate prognosis; requires ≥10% of each component by WHO criteria

Biopsy Concordance with Final Surgical Histology

A landmark 759-patient study established that biopsy technique affects concordance with final surgical histology:

  • CT-guided core biopsy — 44% concordance
  • VATS-guided biopsy — 74% concordance
  • Open surgical biopsy — 83% concordance
  • Concordance reached 100% when ≥10 tissue blocks were sampled

A single 1 cm biopsy represents only 0.1% of total tumor burden (~100–1,000 cm³), explaining underestimation of the biphasic component.

Treatment Implications by Subtype (ASCO 2025)

  • Non-epithelioid (sarcomatoid / biphasic) — ipilimumab plus nivolumab should be offered as the recommended first-line treatment (ASCO Rec. 3.2), and chemotherapy should not be offered unless there are medical contraindications to immunotherapy (Rec. 4.2).[10]
  • Sarcomatoid with ASS1 deficiency (~50% of cases) — may benefit from pegargiminase (ADI-PEG 20) + chemotherapy (ASCO 2025 Conditional Recommendation 4.4).
  • Sarcomatoid histology — patients with histologically confirmed sarcomatoid mesothelioma should not be offered maximal surgical cytoreduction (ASCO 2025 Recommendation 2.3, Strong).[10]

What Is the Peritoneal Cancer Index?

The Peritoneal Cancer Index (PCI) (Sugarbaker system, 0–39 points across 13 abdominal regions) is the standard staging tool for peritoneal mesothelioma. A PCI score ≤20 is generally considered the threshold for cytoreductive surgery (CRS) + HIPEC candidacy; higher scores are associated with incomplete resection and reduced survival benefit. There is no formal TNM staging system specifically validated for peritoneal mesothelioma; the PCI functional score is used clinically as a staging surrogate.

Are There Compensation Implications of Diagnostic Delays?

Mesothelioma's average diagnostic delay of 3–6 months — driven by symptom overlap with benign asbestos-related conditions and the difficulty of distinguishing reactive mesothelial proliferation from early malignancy — has direct legal and financial consequences. Delayed diagnosis can affect asbestos trust fund filing eligibility, statute-of-limitations deadlines for product-liability claims, and a patient's window for treatment trial enrollment. Diagnostic precision matters not only for treatment selection but for legal documentation: histologic subtype, BAP1/CDKN2A status, and exposure history together build the evidentiary record. For specific filing windows, see Mesothelioma_Statute_of_Limitations.

Patients facing a suspected or confirmed mesothelioma diagnosis should pursue both an experienced mesothelioma specialist and a qualified mesothelioma attorney early — diagnostic milestones and compensation deadlines run in parallel, not sequentially.

Frequently Asked Questions

What is the most accurate test to diagnose mesothelioma?

Tissue biopsy with immunohistochemistry remains the gold standard for diagnosing mesothelioma. The most diagnostically powerful combination in 2025–2026 is BAP1 IHC + MTAP IHC (or BAP1 IHC + CDKN2A FISH) — loss of either marker has virtually 100% specificity for malignancy in the mesothelial lineage. VATS-guided biopsy delivers 92–95% sensitivity and is the surgical gold standard; CT-guided core needle biopsy achieves 93% diagnostic yield. Cytology alone is non-diagnostic for sarcomatoid mesothelioma (sensitivity ~6%) and limited for epithelioid (~35%).

Is there a blood test for mesothelioma?

The MESOMARK assay (soluble mesothelin-related peptides, SMRP) is the only FDA-approved serum biomarker, but it is approved for monitoring established mesothelioma — not for primary diagnosis or screening. In an individual patient data meta-analysis, pooled sensitivity was 47% at a specificity of 96% (study ranges 19–68% and 88–100%).[9] IMPRESS tissue methylation profiling has been evaluated in a research validation study.[2]

What changed in mesothelioma staging in 2025?

The IASLC TNM 9th edition, published January 2025, replaced the 8th edition's anatomically defined T1–T4 invasion descriptors with pleural thickness sum (Psum) measured at three axial CT levels (apex, hilum, base). The change was driven by superior inter-observer reproducibility and continuous prognostic value. New T-category boundaries: T1 ≤12 mm Psum (median OS 49.8 mo), T2 12–30 mm (27.5 mo), T3 >31 mm (21.1 mo). The N-descriptor was largely retained. The 8th edition stage groupings remain in widespread clinical use during the 2025–2026 transition.

Should everyone with mesothelioma get genetic testing?

The two guidelines that address it do not fully agree. ASCO's 2025 guideline update, Recommendation 7.1, states that all patients with mesothelioma should be offered germline testing (evidence quality High; strength of recommendation Strong).[3] The European BAP1-TPDS consensus advises that testing "should not routinely be offered" in isolated mesothelioma, while allowing it to be considered on age of onset and other factors (95% consensus, Moderate /weak).[4] The same consensus records that the prevalence of germline BAP1 pathogenic variants in individuals with isolated mesothelioma is low (less than 3%), and that there is a clear association with asbestos exposure.[4] The value of the test is prospective: a positive result qualifies first-degree relatives for cascade screening and surveillance, and can inform treatment discussions. It does not identify the cause of the tested patient's mesothelioma, which remains asbestos exposure.[16]

Is AI accurate for diagnosing mesothelioma?

No AI tool has received FDA approval specifically for mesothelioma diagnosis as of May 2026. Research-grade deep-learning models have demonstrated strong performance: SpindleMesoNET (Naso et al., 2021) achieved AUCs of 0.932, 0.925 and 0.989 on its cross-validation, referral and external test sets, and 92.5% accuracy on the referral set for distinguishing sarcomatoid mesothelioma from benign spindle-cell proliferation, comparable to the 91.7% average accuracy of expert pathologists.[12] MesoNet (Courtiol et al., 2019) improves overall survival prediction from whole-slide images without pathologist annotation.[11] Paige PanCancer Detect received FDA Breakthrough Device Designation in April 2025 for multi-site cancer detection but is not yet cleared and is not mesothelioma-specific.

What is the difference between BAP1 IHC and BAP1 germline testing?

BAP1 IHC is a test performed on biopsy or cytology specimens to look for loss of nuclear BAP1 staining in the tumor; BAP1 loss is a highly specific marker of malignancy in a mesothelial lesion, though caution is warranted on cytology specimens.[1] BAP1 germline testing is a blood-based DNA test looking for inherited pathogenic variants in the BAP1 gene that confer BAP1 Tumor Predisposition Syndrome. The two tests answer different questions: BAP1 IHC asks "is this tissue malignant?"; BAP1 germline testing asks "does this patient have a hereditary cancer-predisposition syndrome?" ASCO 2025 recommends that all patients with mesothelioma be offered germline testing (Rec. 7.1)[3], and BAP1 and MTAP IHC in the appropriate clinical setting when mesothelioma in situ is in the differential (Rec. 6.6).[10]

What is well-differentiated papillary mesothelial tumor (WDPMT)?

WDPMT was reclassified by the WHO 5th Edition Classification of Thoracic Tumours (2021) as a non-malignant entity, distinct from diffuse pleural mesothelioma. WDPMT carries TRAF7 and CDC42 mutations (rather than BAP1 / CDKN2A alterations), which provides molecular confirmation of its non-malignant biology. Distinguishing WDPMT from diffuse malignant mesothelioma is clinically critical — confirmed WDPMT should not receive chemotherapy.

Why does mesothelioma take so long to diagnose?

Diagnostic delay in mesothelioma — typically 3–6 months between symptom onset and confirmed diagnosis — reflects several converging factors. First, presenting symptoms (cough, dyspnea, pleural effusion, weight loss) overlap with common benign asbestos-related conditions and other malignancies. Second, distinguishing early or focal mesothelioma from reactive mesothelial proliferation requires specific IHC (BAP1, MTAP) and FISH (CDKN2A) panels that are not universally available outside expert centers. Third, sarcomatoid mesothelioma is essentially non-diagnostic on pleural fluid cytology (sensitivity ~6%), requiring tissue biopsy. The mean latency from asbestos exposure to mesothelioma diagnosis is 30–50 years, which often makes the exposure history non-obvious at presentation.

References

  1. ↑ 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 Chapel DB, Schulte JJ, Husain AN, Krausz T. Application of immunohistochemistry in diagnosis and management of malignant mesothelioma. Translational Lung Cancer Research. 2020. PMID: 32206567.
  2. ↑ 2.0 2.1 2.2 2.3 Vandenhoeck J, et al. Methylation biomarkers can distinguish pleural mesothelioma from healthy pleura and other pleural pathologies. Molecular Oncology. 2026 Apr. PMID: 41239415.
  3. ↑ 3.0 3.1 3.2 3.3 3.4 Kindler HL, et al. Treatment of Pleural Mesothelioma: ASCO Guideline Update. Journal of Clinical Oncology. 2025 Mar 10. PMID: 39778125.
  4. ↑ 4.0 4.1 4.2 4.3 4.4 4.5 Lalloo F, Kulkarni A, Chau C, Nielsen M, Sheaff M, Steele J, et al. Clinical practice guidelines for the diagnosis and surveillance of BAP1 tumour predisposition syndrome. Eur J Hum Genet. 2023;31(11):1261-1269. PMID 37607989. PMC10620132. PubMed
  5. ↑ 5.0 5.1 5.2 5.3 5.4 Roche. Roche strengthens cancer diagnostic portfolio with launch of test for differential diagnosis of mesothelioma. Media release, Basel. October 28, 2026.
  6. ↑ 6.0 6.1 U.S. Food and Drug Administration. Establishment Registration & Device Listing: VENTANA BAP1 (SP527) Antibody. Accessed October 31, 2026.
  7. ↑ 7.0 7.1 21 CFR 864.1860, Immunohistochemistry reagents and kits. Electronic Code of Federal Regulations.
  8. ↑ 8.0 8.1 21 CFR 807.39, Misbranding by reference to establishment registration or to registration number. Electronic Code of Federal Regulations.
  9. ↑ 9.0 9.1 9.2 Hollevoet K, et al. Serum mesothelin for diagnosing malignant pleural mesothelioma: an individual patient data meta-analysis. Journal of Clinical Oncology. 2012 May 1. PMID: 22412141.
  10. ↑ 10.0 10.1 10.2 10.3 10.4 10.5 American Society of Clinical Oncology; Guideline Central. Treatment of Pleural Mesothelioma: ASCO Guideline Update, guideline summary (Recommendations 2.3, 3.2, 4.2, 6.6, 6.7, 6.8).
  11. ↑ 11.0 11.1 11.2 Courtiol P, et al. Deep learning-based classification of mesothelioma improves prediction of patient outcome. Nature Medicine. 2019 Oct. PMID: 31591589.
  12. ↑ 12.0 12.1 12.2 Naso J, Levine A, Farahani H, Chirieac LR, Dacic S, Wright JL, Lai CK, Yang H, Jones SJM, Bashashati A, Yip S, Churg A. Deep-learning based classification distinguishes sarcomatoid malignant mesotheliomas from benign spindle cell mesothelial proliferations. Modern Pathology. 2021 Nov. PMID: 34112957.
  13. ↑ 13.0 13.1 13.2 British Thoracic Society Pleural Disease Guidelines — Online Appendix D1: Best Imaging Modality for Pleural Malignancy. BTS, 2023.
  14. ↑ Robinson BW, et al. Mesothelin-family proteins and diagnosis of mesothelioma. The Lancet. 2003 Nov 15. PMID: 14630441.
  15. ↑ Pass HI, et al. Fibulin-3 as a blood and effusion biomarker for pleural mesothelioma. New England Journal of Medicine. 2012 Oct 11. PMID: 23050525.
  16. ↑ 16.0 16.1 16.2 16.3 16.4 International Agency for Research on Cancer. Asbestos (Chrysotile, Amosite, Crocidolite, Tremolite, Actinolite, and Anthophyllite). IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Volume 100C: Arsenic, Metals, Fibres, and Dusts. Lyon: IARC; 2012. Section 5, Evaluation: "Asbestos causes mesothelioma and cancer of the lung, larynx, and ovary."
  17. ↑ Rafizadeh M, Bodd F, Sauter J, et al. Molecular profiling of primary tumors and distant metastases in malignant mesotheliomas. Abstract 4162P, ESMO Congress 2026, Madrid, October 2026. Funded by the National Institutes of Health/National Cancer Institute. Conference abstract; not a peer-reviewed paper.
  18. ↑ 18.0 18.1 Pilarski R, Byrne L, Carlo MI, et al. BAP1 Tumor Predisposition Syndrome. GeneReviews. Adam MP, et al., editors. Seattle (WA): University of Washington, Seattle; 2016 Oct 13 [updated 2024 Dec 5]. PMID: 27748099
  19. ↑ 19.0 19.1 19.2 Herman EJ, Debernardi C, Hafeez KS, et al. Rare germline variants and genetic susceptibility to pleural mesothelioma: moving beyond DNA repair. medRxiv preprint, posted October 2, 2026. doi:10.64898/2026.09.29.26364318. Preprint; not peer-reviewed.
  20. ↑ See Mesothelioma_Biopsy_Procedures for diagnostic yield by biopsy technique.