Skip to content
Bending the Curve Trial by Trial
?

Pleura and peritoneum (serosal membranes of the chest and abdomen; thoracic tumours)

Malignant mesothelioma

How asbestos, staging, surgery, chemotherapy and immunotherapy have been studied in malignant pleural and peritoneal mesothelioma.

5-year relative survival · United States, SEER 9 registries (SEER Cancer Statistics Review 1975-2017)

9.7%→10.7%

Diagnosed 1975-1977 to Diagnosed 2010-2016. Read these with the caveats under “Then and now”.Source

24 milestones, by decade

  • The Health and Safety Executive counted 2,146 mesothelioma deaths in Great Britain in 2024 (a provisional figure), below the average of 2,508 a year for 2011 to 2020. It reports that annual deaths stayed broadly level during the 2010s at around 2,500, around 10 times the annual number of the early 1970s, and gives a latency of typically at least 30 years, so most deaths today follow asbestos exposures that occurred during 1950 to 1980.
  • In the SEER 9 registries, 5-year relative survival for mesothelioma was 9.7% for people of both sexes diagnosed in 1975 to 1977 and 10.7% for those diagnosed in 2010 to 2016. In the SEER Cancer Statistics Review table only the male figures (3.0% and 8.3%) carry the mark for a statistically significant difference; the female figures are 26.7% and 18.6%, and the table does not split pleural from peritoneal disease. SEER has no Cancer Stat Facts page for mesothelioma, so these figures come from its archived review.
  • FDA approval letters for malignant pleural mesothelioma are dated 4 February 2004 (pemetrexed), 2 October 2020 (nivolumab with ipilimumab, one letter for each drug) and 17 September 2024 (pembrolizumab with pemetrexed and platinum chemotherapy). In the EU the pembrolizumab indication (Commission decision 4 April 2025) is limited to non-epithelioid disease.

Evidence current to October 2026 · Checked against primary sources to the extent possible

A guided tour of 12 milestones. You can pause or step through it yourself.

Filter milestonesshowing all 24
Treatment type
Setting

Chemotherapy

  1. 1960 · Understanding · UnderstandingPleural mesothelioma and probable crocidolite exposure (Wagner, 1960)In 1960 a paper described thirty-three people with diffuse pleural mesothelioma, an uncommon tumour of the chest lining, in the North Western Cape Province of South Africa. All but one had probably been exposed to crocidolite, or Cape blue, asbestos, mostly in the Asbestos Hills. It was a case series; its abstract gives no risk estimate.Wagner, Sleggs and Marchand, Br J Ind Med 1960: case series of 33 diffuse pleural mesotheliomas (22 men, 11 women, ages 31 to 68). All but one had probable exposure to crocidolite (Cape blue) asbestos, most in the Asbestos Hills west of Kimberley, North Western Cape Province; the tumour was described as rarely seen elsewhere in South Africa. The abstract reports exposure as probable and gives no comparison group or risk estimate.
  2. 1965 · Understanding · UnderstandingLondon series of 83 mesotheliomas: asbestos exposure histories (1965)In 1965 a London hospital series of eighty-three people with mesothelioma looked at their asbestos exposure. About half of those with full histories reported work or household exposure, compared with about one in eight hospital patients with other diseases. A New England Journal of Medicine paper on the relation between asbestos and mesothelioma appeared the same year.Newhouse and Thompson, Br J Ind Med 1965: 83 London Hospital patients with necropsy- or biopsy-confirmed mesothelioma (41 men, 42 women; 56 pleural, 27 peritoneal). Of 76 with full histories, 40 (52.6%) reported occupational or domestic asbestos exposure versus 9 (11.8%) of 76 hospital patients with other diseases (p < 0.001). First exposure to illness: 16 to 55 years. Selikoff, Churg and Hammond (N Engl J Med 1965) is cited by title only.

Molecular origins

  1. 1995 · Understanding · UnderstandingIMIG TNM staging system for pleural mesothelioma (1995)In 1995 the International Mesothelioma Interest Group published a TNM-based staging system for malignant pleural mesothelioma, developed at a consensus meeting of researchers. It gave precise anatomical definitions of the tumour, in contrast to five earlier systems. A later overview says it was accepted by the AJCC and UICC and became the international staging standard.Rusch, Chest 1995, from the IMIG: TNM-based system from a consensus meeting that included originators of earlier systems. T1 to T4 give anatomic definitions; N0 to N3 are virtually identical to the lung cancer N descriptors. T1 and N0 tumours fall in stages I and II, node-positive tumours in stage III, and T4, N3 or M1 disease in stage IV. The abstract compares it with five previous systems and reports no survival figures.

Targeted therapy

  1. 1999 · Treatment · SurgeryTrimodality therapy with extrapleural pneumonectomy: 183 patientsA US surgical group reported 183 people with pleural mesothelioma treated in 1980-1997 with extrapleural pneumonectomy, then chemotherapy and radiotherapy. Of 176 who survived the operative period, 38% were alive at two years and 15% at five. Epithelial type, clear margins and uninvolved nodes were linked to longer survival. There was no comparison group.Series of 183 patients, 1980-1997: extrapleural pneumonectomy then adjuvant chemotherapy and radiotherapy. Perioperative mortality 3.8% (7 deaths), morbidity 50%; median follow-up 13 months. In 176 who survived the perioperative period: 2- and 5-year survival 38% and 15%, median 19 months. Univariate: epithelial type (median 26 months), negative margins (23), node-negative extrapleural nodes (21). Cox relative risk of death, nonepithelial type: 3.0 (CI 2.0-4.5). Uncontrolled.
  2. 2003 · Treatment · ChemotherapyPemetrexed plus cisplatin versus cisplatin in pleural mesotheliomaIn a trial of people with pleural mesothelioma who could not have curative surgery, adding pemetrexed to cisplatin extended median survival from about nine months to about twelve. Tumours shrank in about four in ten people given the combination and about one in six given cisplatin alone. Vitamin supplements were added to reduce side effects.Randomised phase 3, 456 assigned (448 treated): pemetrexed 500 mg/m2 plus cisplatin 75 mg/m2 versus cisplatin alone, every 21 days, in chemotherapy-naive patients not eligible for curative surgery. Median survival 12.1 versus 9.3 months (log-rank P = 0.020; HR 0.77, no CI in the abstract). Median TTP 5.7 versus 3.9 months (P = 0.001); response rate 41.3% versus 16.7% (P < 0.0001). Folic acid and vitamin B12 were added after 117 patients had enrolled.
  3. 2009 · Treatment · SurgeryPeritoneal mesothelioma: cytoreductive surgery with HIPECA registry of 405 people with diffuse peritoneal mesothelioma, treated at several centres with surgery to remove visible tumour (92% also had heated chemotherapy inside the abdomen), reported median survival of 53 months, with 60% alive at three years and 47% at five. It describes selected patients and was not a randomised comparison.Multi-institutional registry, 405 patients with diffuse malignant peritoneal mesothelioma (79% epithelial; mean age 50): cytoreductive surgery with HIPEC in 372 (92%). Follow-up complete in 401. Median OS 53 months (1-235 as printed); 3- and 5-year survival 60% and 47%. Grade 3-4 complications 31% (127); 9 perioperative deaths (2%). Epithelial subtype, no nodal metastasis, CC-0 or CC-1 and HIPEC were associated with longer survival in multivariate analysis. Observational.

Immunotherapy

  1. 2011 · Understanding · UnderstandingBAP1 mutations: germline predisposition and somatic loss (2011)Two 2011 Nature Genetics papers reported changes in the BAP1 gene in mesothelioma. One found disabling BAP1 mutations, acquired in the tumour, in nearly a quarter of pleural mesotheliomas. The other found inherited BAP1 mutations in two families with a high rate of mesothelioma, some members also developing eye melanoma, and described a BAP1-related cancer syndrome.Bott 2011 (Nat Genet): integrated genomic analysis of 53 MPM tumour samples guided sequencing and found somatic inactivating BAP1 mutations in 23% of MPMs. Testa 2011 (Nat Genet): germline BAP1 mutations in two families with a high incidence of mesothelioma, some carriers also developing uveal melanoma; germline BAP1 mutations in 2 of 26 sporadic mesotheliomas (both with prior uveal melanoma). The abstracts give no penetrance or risk estimates.
  2. 2011 · Reversal · SurgeryMARS: extrapleural pneumonectomy versus no EPP, a feasibility studyMARS was a small randomised feasibility study in the UK. After chemotherapy, 50 people with pleural mesothelioma were assigned to a major operation (extrapleural pneumonectomy) followed by radiotherapy, or to no operation. Median survival was 14.4 months in the operation group and 19.5 months without it, from small numbers. The authors concluded that a larger trial was not feasible.Randomised feasibility study, 12 UK hospitals: after induction platinum chemotherapy, 50 of 112 registered patients were randomised to EPP then hemithorax radiotherapy (24) or no EPP (26). OS HR for EPP versus no EPP 1.90 (95% CI 0.92-3.93; exact p=0.082); adjusted HR 2.75 (1.21-6.26; p=0.016). Median survival 14.4 versus 19.5 months. Serious adverse events: ten with EPP, two without.
  3. 2015 · Treatment · TargetedMAPS: bevacizumab added to cisplatin-pemetrexedIn a French trial of people with untreated pleural mesothelioma, adding bevacizumab, a drug that blocks tumour blood-vessel growth, to cisplatin and pemetrexed extended median survival from about sixteen months to about nineteen. High blood pressure and blood clots were more frequent with bevacizumab.Open-label phase 3 (IFCT-GFPC-0701), 448 randomised (223 to bevacizumab 15 mg/kg plus cisplatin 75 mg/m2 and pemetrexed 500 mg/m2, 225 to chemotherapy alone), 21-day cycles for up to six cycles. Median OS 18.8 versus 16.1 months, HR 0.77 (95% CI 0.62-0.95), p=0.0167 (ITT). Grade 3 or higher hypertension 23% versus 0; thrombotic events 6% versus 1%.
  4. 2015 · Reversal · TargetedVorinostat, NGR-hTNF, nintedanib, anetumab: no superiority shownFour randomised trials tested other drugs in pleural mesothelioma. Vorinostat, NGR-hTNF and anetumab ravtansine were tested after chemotherapy; nintedanib was added to first-line chemotherapy. None met its main goal of improving survival (two trials) or progression-free survival (two trials) over placebo or the comparison treatment. An earlier, smaller nintedanib study had looked more favourable.VANTAGE-014 (vorinostat versus placebo, n=661): median OS 30.7 versus 27.1 weeks, HR 0.98 (0.83-1.17). NGR015 (NGR-hTNF versus placebo, each with investigator-choice therapy, n=400): OS 8.5 versus 8.0 months, HR 0.94 (0.75-1.18). LUME-Meso phase 3 (first line, n=458): PFS 6.8 versus 7.0 months, HR 1.01 (0.79-1.30). ARCS-M (anetumab ravtansine versus vinorelbine, n=248): PFS 4.3 versus 4.5 months, HR 1.22 (0.85-1.74). Not pooled.
  5. 2015 · Treatment · ChemotherapySAKK 17/04: induction chemotherapy, EPP, then radiotherapy or notSAKK 17/04 gave 151 people with pleural mesothelioma chemotherapy, then planned extrapleural pneumonectomy. Of 96 with a complete macroscopic resection, 54 were randomised: 27 to high-dose chest radiotherapy and 27 to none. Median time free of chest relapse, from surgery, was 9.4 months with radiotherapy and 7.6 without. The authors did not support routine radiotherapy.Two-part phase 2, 14 hospitals. Part 1 (single arm): three cycles of cisplatin-pemetrexed then extrapleural pneumonectomy; 151 analysed, 113 (75%) had EPP, complete macroscopic resection in 96 (64%). Part 2: 54 patients with complete macroscopic resection randomised 27:27 to high-dose hemithoracic radiotherapy or none; median locoregional relapse-free survival (ITT) 9.4 months (95% CI 6.5-11.9) with radiotherapy versus 7.6 (4.5-10.7) without. One patient died of pneumonitis.
  6. 2016 · Understanding · UnderstandingGenomic landscape of pleural mesothelioma: molecular subtypesTwo large genomic studies, in 2016 and 2018, mapped pleural mesothelioma. The first analysed more than two hundred tumours and described four molecular subtypes and recurrently altered genes such as BAP1 and NF2. The second, from The Cancer Genome Atlas, studied seventy-four tumours, described a rare subtype with TP53 and SETDB1 mutations and reported strong VISTA expression in epithelioid disease.Bueno 2016 (Nat Genet): transcriptomes (n = 211), whole exomes (n = 99) and targeted exomes (n = 103) from 216 MPM tumours; four RNA-seq subtypes (sarcomatoid, epithelioid, biphasic-epithelioid, biphasic-sarcomatoid); significantly mutated genes include BAP1, NF2, TP53 and SETD2. Hmeljak 2018 (Cancer Discov, TCGA): 74 MPMs; histology-independent prognostic subsets; a subtype with TP53 and SETDB1 mutations and extensive loss of heterozygosity; strong VISTA expression in epithelioid MPM.
  7. 2016 · Understanding · UnderstandingIASLC mesothelioma staging project: proposals for the eighth editionIn 2016 the International Association for the Study of Lung Cancer reported analyses of a database of more than three thousand pleural mesothelioma cases from nearly thirty centres. They proposed merging some T categories and the nodal categories, and new stage groupings, for the eighth edition of the TNM classification. The authors call it the first evidence-based revision.IASLC database, 29 centres, 3519 cases submitted. Rusch: 2414 analyzable, 84 clinical M1; median OS 9.7 months in cM1 versus 13.4 in locally advanced cM0 (p = 0.0013). Proposals: collapse T1a and T1b into T1 (Nowak); one ipsilateral intrathoracic N1 category, former N3 becomes N2 (Rice); stages IA T1N0, IB T2-3N0, II T1-2N1, IIIA T3N1, IIIB T1-3N2 or any T4, IV any M1.
  8. 2016 · Reversal · RadiationProphylactic radiotherapy to procedure tracts, 1995 to SMARTMesothelioma can grow along the track of a biopsy or drain, so radiotherapy to these sites has been tried as prevention. A small 1995 trial of 40 people found tract metastases in 8 of 20 without radiotherapy and none in 20 with it. Three later randomised trials, the largest in 203 people, found no significant difference.SMART (UK, 22 hospitals): 203 randomised to immediate (102) or deferred (101) radiotherapy, 21 Gy in three fractions; procedure-tract metastasis within 7 cm at 12 months in 9 (9%) versus 16 (16%), OR 0.51 (95% CI 0.19-1.32; p=0.14). Boutin 1995: 40 patients, none of 20 versus 8 of 20 (40%). Bydder 2004 (58 sites) and O'Rourke 2007 (61 patients): no significant difference. Units, doses and designs differ; not pooled.
  9. 2016 · Understanding · UnderstandingWHO classification of pleural tumours: 2015 and 2021 editionsThe World Health Organization revised its classification of pleural tumours in 2015 and again in 2021. Journal summaries report that the three main mesothelioma subtypes (epithelioid, biphasic and sarcomatoid) stayed the same. The 2021 edition added mesothelioma in situ, introduced nuclear grading and dropped the word malignant from the names of diffuse and localised mesothelioma.2015 edition (Galateau-Salle 2016): histologic classification of pleural malignant mesothelioma unchanged from 2004; pleomorphic epithelioid subtype recognised as poor-prognosis, like sarcomatoid. 2021 edition (Sauter 2022): pleura and pericardium in one chapter; 'malignant' prefix dropped; mesothelioma in situ added (BAP1/MTAP loss by IHC or CDKN2A homozygous deletion by FISH); nuclear grading for epithelioid diffuse mesothelioma; BAP1, EZH2, MTAP IHC useful against benign proliferations.
  10. 2017 · Reversal · ImmunotherapyDETERMINE: tremelimumab did not improve survival over placeboIn 571 people whose mesothelioma had progressed after one or two earlier treatments, the immunotherapy drug tremelimumab was compared with placebo. Median survival was 7.7 months with tremelimumab and 7.3 months with placebo, not a significant difference. Severe side effects were more common with tremelimumab (65% against 48%).DETERMINE, randomised (2:1) double-blind placebo-controlled phase 2b: tremelimumab 10 mg/kg every 4 weeks for 7 doses, then every 12 weeks, in pleural or peritoneal mesothelioma after one or two systemic regimens (n=571). Primary endpoint OS (ITT): median 7.7 months (95% CI 6.8-8.9) versus 7.3 (5.9-8.7); HR 0.92 (0.76-1.12), p=0.41. Grade 3 or worse adverse events 65% versus 48%.
  11. 2019 · Treatment · ImmunotherapyPD-1 blockade after chemotherapy: MAPS2, PROMISE-meso and CONFIRMThree separate trials tested PD-1 immunotherapy after chemotherapy. MAPS2 (125 people, no comparison between arms) saw disease control at 12 weeks in 40% on nivolumab. PROMISE-meso (144) found no survival gain over chemotherapy. CONFIRM (332) favoured nivolumab over placebo at first, but its final survival comparison was not statistically significant.MAPS2 (randomised non-comparative phase 2, n=125): 12-week disease control 40% (95% CI 28-52) with nivolumab, 52% (39-64) with ipilimumab added (ITT). PROMISE-meso (phase 3 versus chemotherapy, n=144): BICR-PFS HR 1.06 (0.73-1.53); OS HR 1.12 (0.74-1.69). CONFIRM (placebo-controlled phase 3, n=332): preliminary OS HR 0.69 (0.52-0.91); final OS 9.49 versus 6.77 months, HR 0.81 (0.64-1.04), p=0.096. Not pooled.
  12. 2019 · Treatment · TargetedTumour treating fields with chemotherapy: STELLAR and the FDA HDEIn a single-group study of 80 people with untreated pleural mesothelioma that could not be removed, a device that applies electric fields to the chest was added to pemetrexed and platinum chemotherapy. Median survival was about 18 months, and there was no comparison group. In May 2019 the FDA approved the device under its Humanitarian Device Exemption pathway.STELLAR (single-arm phase 2, 12 European sites): TTFields 150 kHz to the thorax with pemetrexed plus cisplatin or carboplatin for up to six cycles, then TTFields maintenance, in untreated unresectable MPM (n=80). Primary endpoint OS (ITT): median 18.2 months (95% CI 12.1-25.8); median follow-up 12.5 months. FDA HDE H180002 approved 23 May 2019: a device pathway, with the basis stated as reasonable assurance of safety and probable benefit.

Precision & engineered

  1. 2020 · Treatment · Immunotherapy⚑ needs reviewCheckMate 743: nivolumab plus ipilimumab versus chemotherapyIn people with untreated pleural mesothelioma that could not be removed by surgery, the immunotherapy pair nivolumab and ipilimumab, given without chemotherapy, was compared with chemotherapy. Median survival was about eighteen months versus about fourteen, and about four in ten were alive at two years versus about one in four. The FDA approved it in October 2020.Open-label phase 3, 605 randomised: nivolumab 3 mg/kg every 2 weeks plus ipilimumab 1 mg/kg every 6 weeks for up to 2 years (n=303) versus platinum-pemetrexed (n=302). Interim analysis (median follow-up 29.7 months): median OS 18.1 versus 14.1 months, HR 0.74 (96.6% CI 0.60-0.91), p=0.0020; 2-year OS 41% versus 27%. Updates: 3-year OS 23% versus 15% (HR 0.73, 95% CI 0.61-0.87); 5-year OS 14% versus 6% (HR 0.74, 95% CI 0.62-0.88).
  2. 2023 · Treatment · ImmunotherapyIND.227 / KEYNOTE-483: pembrolizumab added to platinum-pemetrexedIn people with untreated pleural mesothelioma, adding the immunotherapy pembrolizumab to platinum-pemetrexed chemotherapy gave a median survival of about seventeen months versus about sixteen with chemotherapy alone. About one in four was alive at three years versus about one in six. The FDA approved the combination without a histology restriction; the European Union limited it to non-epithelioid disease.Open-label phase 3 part of a phase 2/3 trial, 440 randomised: cisplatin or carboplatin with pemetrexed plus pembrolizumab 200 mg every 3 weeks (n=222) or alone (n=218). Final analysis (median follow-up 16.2 months): median OS 17.3 versus 16.1 months, HR 0.79 (95% CI 0.64-0.98), two-sided p=0.0324; 3-year OS 25% versus 17%. Subgroups (EMA): OS HR 0.89 epithelioid, 0.57 non-epithelioid.
  3. 2024 · Treatment · ChemotherapyATOMIC-Meso: pegargiminase with chemotherapy, non-epithelioid diseaseIn a double-blind trial of about 250 people with untreated non-epithelioid pleural mesothelioma, adding the arginine-depleting enzyme pegargiminase to platinum-pemetrexed chemotherapy gave a median survival of about nine months versus about eight months with placebo plus chemotherapy. Severe side effects were more frequent with pegargiminase. No approval for this use was found.Double-blind phase 2/3 (43 centres, 5 countries), 249 randomised 1:1 (125 and 124) to weekly intramuscular pegargiminase or placebo with pemetrexed and cisplatin or carboplatin for up to 6 cycles; non-epithelioid pleural mesothelioma only. Primary endpoint OS (all randomised): median 9.3 months (95% CI 7.9-11.8) versus 7.7 (6.1-9.5), HR 0.71 (95% CI 0.55-0.93), P=0.02. PFS 6.2 versus 5.6 months, HR 0.65 (95% CI 0.46-0.90). Grade 3 or higher related adverse events 28.8% versus 16.9%.
  4. 2024 · Reversal · SurgeryMARS 2: extended pleurectomy decortication plus chemotherapyMARS 2 randomised 335 people in the UK with pleural mesothelioma judged operable. After two cycles of chemotherapy, about half were assigned to an extended operation (pleurectomy decortication) plus more chemotherapy and about half to chemotherapy alone. Median survival was shorter with surgery (19.3 versus 24.8 months), and serious side effects were more common.Phase 3 pragmatic open-label RCT, 26 UK hospitals; 335 randomised after two chemotherapy cycles (169 surgery plus chemotherapy, 166 chemotherapy alone); primary outcome OS, ITT. Median OS 19.3 (IQR 10.0-33.7) versus 24.8 months (12.6-37.4); restricted mean survival time difference at 2 years -1.9 months (95% CI -3.4 to -0.3; p=0.019). Grade 3 or worse serious adverse events (events, not participants) 318 versus 169 (IRR 3.6; 95% CI 2.3-5.5).
  5. 2025 · Reversal · ImmunotherapyBEAT-meso: atezolizumab added to bevacizumab and chemotherapyIn a trial of 400 people with untreated pleural mesothelioma, adding the immunotherapy atezolizumab to bevacizumab and chemotherapy delayed progression by about a month and a half and raised median survival by about two months, but the survival difference was not statistically significant, so the main goal was missed. Non-epithelioid disease appeared to gain more.Open-label phase 3 (ETOP 13-18), 400 randomised 1:1: atezolizumab 1200 mg plus bevacizumab 15 mg/kg plus carboplatin-pemetrexed (ABC) versus bevacizumab plus carboplatin-pemetrexed (BC). Primary endpoint OS in all randomised patients not met: median 20.5 versus 18.1 months, HR 0.84 (95% CI 0.66-1.06), P=0.14. Median PFS 9.2 versus 7.6 months, HR 0.72 (95% CI 0.59-0.89), P=0.0021. OS HR 0.51 non-epithelioid, 1.01 epithelioid (interaction P=0.012).
  6. 2025 · Treatment · ImmunotherapyPerioperative nivolumab, with or without ipilimumab: phase 2 trialA US phase 2 trial gave 30 people with operable pleural mesothelioma nivolumab (16 people) or nivolumab with ipilimumab (14 people) before surgery. Groups were enrolled one after the other, not randomly. Most in each group reached surgery on schedule (81% and 86%), with one dose-limiting side effect per group. Survival was exploratory and the groups were not compared.Non-randomised phase 2, sequential arms, resectable epithelioid or biphasic pleural mesothelioma (30 patients): Arm A nivolumab (16), Arm B nivolumab plus ipilimumab (14), then surgery, optional chemotherapy or radiotherapy and adjuvant nivolumab. Co-primary feasibility and safety: proceeded to surgery 81.3% and 85.7%; one DLT per arm (6.3%, 7.1%). Exploratory median PFS 9.6 and 19.8 months, OS 19.3 and 28.6. Not powered for efficacy; no between-arm comparison.

Then and now

  • 5-year relative survival · United States, SEER 9 registries (SEER Cancer Statistics Review 1975-2017)

    • 9.7%

      Diagnosed 1975-1977

    • 10.7%

      Diagnosed 2010-2016

    People of all races and both sexes combined (the 'Total' column of Table 17.4, mesothelioma) diagnosed in the nine SEER 9 areas (San Francisco, Connecticut, Detroit, Hawaii, Iowa, New Mexico, Seattle, Utah and Atlanta), with follow-up into 2017. The row states no age range or number of people, and the table does not split pleural from peritoneal disease.

    SEER's own review tables, not a Cancer Stat Facts figure (SEER has no Cancer Stat Facts page for mesothelioma). Relative survival compares survival with that expected in the general population; the table footnote says expected survival for these rows is derived from the U.S. Annual Life Tables. In the table, the footnote marking the difference between 1975-1977 and 2010-2016 as statistically significant (p<.05) is attached to the male figures only (3.0% and 8.3%), not to the both-sexes figures; the female figures in the same table are 26.7% and 18.6%, and the table flags no standard error of 5 percentage points or more for this both-sexes row. The table does not describe changes in case mix, coding or diagnosis over the decades, which can affect a comparison of periods. SEER no longer produces the Cancer Statistics Review (not since the November 2021 data submission), and the 1975-2018 edition has no mesothelioma section, so 2010-2016 is the latest period of this series. Do not compare these values with the American Cancer Society figures for 2015-2021 below, which cover pleural mesothelioma only, a different period and a registry set that the American Cancer Society page does not name.

    Source

  • 5-year relative survival · United States (American Cancer Society)

    • 15%

      Diagnosed 2015-2021, all SEER stages combined

    People diagnosed with pleural mesothelioma between 2015 and 2021 in the United States, all SEER stages combined, as printed on the American Cancer Society page 'Survival Rates for Pleural Mesothelioma' (last revised 3 February 2026). The page states that peritoneal and pericardial mesothelioma are not covered and gives no age range, sex or number of people.

    Secondary source: the American Cancer Society says it relies on the NCI SEER database; its references are the SEER Summary Stage 2018 coding instructions and SEER*Explorer (accessed 13 June 2025), and it does not name the registry set. SEER has no Cancer Stat Facts page for mesothelioma, so this is not a Cancer Stat Facts figure. By the page's own list, the figures apply to stage at first diagnosis only and do not account for age, overall health, mesothelioma subtype, whether the tumour can be removed or response to treatment; the page adds that people diagnosed more recently may have a better outlook, because the figures describe people diagnosed and treated at least 5 years earlier. The page prints 15% both for regional stage and for all SEER stages combined; this was recorded as printed and not checked against SEER*Explorer. This is a different series from the SEER 9 figures for 1975-1977 and 2010-2016, so no change over time is implied between them.

    Source

  • 5-year relative survival · United States (American Cancer Society)

    • 23%

      Diagnosed 2015-2021, localized stage

    • 15%

      Diagnosed 2015-2021, regional stage

    • 11%

      Diagnosed 2015-2021, distant stage

    Pleural mesothelioma by SEER stage at diagnosis, for people diagnosed between 2015 and 2021 in the United States, as printed on the American Cancer Society page 'Survival Rates for Pleural Mesothelioma'. In the page's wording localized means limited to the pleura, regional means spread to nearby structures or lymph nodes, and distant means spread to distant sites such as the abdomen, bones or the pleura on the opposite side of the body; the page states that SEER does not use AJCC TNM stages. Pleural disease only; no age range, sex or number of people is given.

    Secondary source: the American Cancer Society says it relies on the NCI SEER database; its references are the SEER Summary Stage 2018 coding instructions and SEER*Explorer (accessed 13 June 2025), and it does not name the registry set. SEER has no Cancer Stat Facts page for mesothelioma, so this is not a Cancer Stat Facts figure. By the page's own list, the figures apply to stage at first diagnosis only and do not account for age, overall health, mesothelioma subtype, whether the tumour can be removed or response to treatment; the page adds that people diagnosed more recently may have a better outlook, because the figures describe people diagnosed and treated at least 5 years earlier. The page prints 15% both for regional stage and for all SEER stages combined; this was recorded as printed and not checked against SEER*Explorer. These are separate figures for three stage groups of the same 2015-2021 group of people, not a change over time, and they are not to be subtracted from one another or from the all-stages figure.

    Source

  • 10-year relative survival · England (Cancer Research UK)

    • 1.9%

      Diagnosed 2013-2017 (predicted, as the page words it)

    People diagnosed with mesothelioma in England, as printed on the Cancer Research UK health professional page 'Mesothelioma survival statistics': '2 in 100 (1.9%) ... survive their disease for ten years or more, it is predicted (2013-2017)'. The page states no age range and gives no split by site (pleural, peritoneal) or stage for this figure.

    The health professional page read does not define the measure. Cancer Research UK's public mesothelioma survival page (last reviewed 18 May 2023) describes its one-year and five-year figures as net survival, survival of people with cancer after taking background mortality into account; whether the same wording applies to this ten-year figure is not stated on the page read. It is a different country and registry from the SEER figures above, so the two are shown separately and not compared. The health professional page gives no earlier mesothelioma figure, so no change over time is available from it. The page words the figure as predicted, so it is not an observed 10-year follow-up of people diagnosed in 2013-2017.

    Source

Survival figures are affected by earlier detection (lead time) and overdiagnosis, and by changes in who is diagnosed. For screened cancers, read them alongside mortality.

Evidence chains

A lane follows one defined population and setting, where each new trial tests against what was then standard. Compare only within a lane, and read every comparison as illustrative.

Unresectable pleural mesothelioma, no earlier treatment: first-line systemic therapy

Adults with previously untreated malignant pleural mesothelioma that was unresectable or not amenable to curative surgery · Metastatic · endpoint: OS

Results are not comparable across trials.

The control arms differ: cisplatin alone (EMPHACIS), cisplatin-pemetrexed (MAPS), platinum-pemetrexed (CheckMate 743 and IND.227) and bevacizumab with carboplatin-pemetrexed (BEAT-meso). The CheckMate 743 experimental arm, nivolumab plus ipilimumab, contained no chemotherapy; the other experimental arms added a drug to a chemotherapy backbone or, in EMPHACIS, added pemetrexed to cisplatin. Eligibility differs: MAPS enrolled people aged 18 to 75 with ECOG performance status 0 to 2 and no substantial cardiovascular comorbidity, CheckMate 743 and IND.227 required ECOG 0 or 1, and EMPHACIS a Karnofsky status of 70 or more. Era, histology mix (BEAT-meso reports 78% epithelioid), follow-up and later therapy also differ. CheckMate 743 is drawn from its prespecified interim analysis and IND.227 from its final analysis, so the survival results are not at the same stage. Reported p-values differ in source and sidedness: for IND.227 the Lancet paper gives a two-sided 0.0324, the FDA label prints 0.0162 without stating its sidedness, and the EMA assessment gives 0.0162 as one-sided; EMPHACIS reports a two-sided log-rank test. ATOMIC-Meso (non-epithelioid disease only) and all subgroup, five-year and phase 2 results are not plotted. Compare only within each trial's own two arms; the lane makes no recommendation and no claim that regimens are equivalent.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

0510152025OS, median (months)20032015202020232025EMPHACISMAPSCheckMate 743IND.227 / KEYNOTE-483BEAT-meso
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS, median, across 5 trials in this lane. EMPHACIS (2003): experimental 12.1 months vs control 9.3 months; MAPS (2015): experimental 18.8 months vs control 16.1 months; CheckMate 743 (2020): experimental 18.1 months vs control 14.1 months; IND.227 / KEYNOTE-483 (2023): experimental 17.3 months vs control 16.1 months; BEAT-meso (2025): experimental 20.5 months vs control 18.1 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
EMPHACIS2003OS9.3 months12.1 months
MAPS2015OS16.1 months18.8 months
CheckMate 7432020OS14.1 months18.1 months
IND.227 / KEYNOTE-4832023OS16.1 months17.3 months
BEAT-meso2025OS18.1 months20.5 months

Mesothelioma that progressed after earlier systemic treatment: later-line therapy

Adults with malignant mesothelioma that had progressed after earlier systemic treatment (pleural only, or pleural and peritoneal, depending on the trial) · Relapsed / refractory · endpoint: OS

Results are not comparable across trials.

The populations differ. NGR015 and the anetumab ravtansine study enrolled pleural mesothelioma only, while DETERMINE and CONFIRM enrolled pleural or peritoneal disease (316 of 332 in CONFIRM had pleural disease). The line of treatment differs: NGR015 enrolled people after one pemetrexed-based regimen, the anetumab study after first-line platinum-pemetrexed with or without bevacizumab, DETERMINE after one or two earlier systemic treatments, and CONFIRM after at least one platinum-based course, with most people in the third-line setting according to its authors. The comparators differ: placebo in DETERMINE and CONFIRM, placebo plus best investigator choice in NGR015, and active vinorelbine in the anetumab study. Overall survival is the primary endpoint in DETERMINE and NGR015 and a co-primary endpoint with progression-free survival in CONFIRM, but a secondary endpoint in the anetumab study, which randomised only people whose tumours overexpressed mesothelin. The CONFIRM entry is its final analysis; its earlier preliminary analysis is a separate result. VANTAGE 014 (medians recorded in weeks), PROMISE-meso (hazard ratio only), MAPS2 (no control arm or overall survival result) and LUME-Meso (a first-line trial) are not drawn. Compare only within each trial's own two arms; the lane makes no recommendation and no claim that agents are equivalent.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

02468101214OS, median (months)2015201520172019NGR015ARCS-MDETERMINECONFIRM
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS, median, across 4 trials in this lane. NGR015 (2015): experimental 8.5 months vs control 8 months; ARCS-M (2015): experimental 9.5 months vs control 11.6 months; DETERMINE (2017): experimental 7.7 months vs control 7.3 months; CONFIRM (2019): experimental 9.49 months vs control 6.77 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
NGR0152015OS8 months8.5 months
ARCS-M2015OS11.6 months9.5 months
DETERMINE2017OS7.3 months7.7 months
CONFIRM2019OS6.77 months9.49 months

Operable pleural mesothelioma: radical surgery after chemotherapy versus no radical surgery

People with pleural mesothelioma judged fit for, or amenable to, radical surgery and randomised after chemotherapy to surgery or no surgery · Perioperative · endpoint: OS

Results are not comparable across trials.

MARS describes itself as a feasibility study: 50 people were randomised, its paper states that a trial powered for overall survival would have needed 670 patients, and survival is counted from randomisation, which followed induction chemotherapy. MARS 2 is a phase 3 trial of 335 people with overall survival as its primary outcome; because the proportional-hazards assumption was violated, its primary comparison was the difference in restricted mean survival time at 2 years, so the plotted result carries no hazard ratio, and the intervals printed beside its medians are interquartile ranges. The operations differ: extrapleural pneumonectomy followed by hemithorax radiotherapy in MARS, and pleurectomy decortication or extended pleurectomy decortication in MARS 2. The chemotherapy also differs in description: MARS records induction platinum-based chemotherapy before randomisation, while MARS 2 records two cycles before randomisation and two to four after, without naming the regimen in this record. The eras differ (randomisation ran from 2005 to 2008 in MARS and from 2015 to 2021 in MARS 2), and 86% of the people in MARS 2 had epithelioid disease. Neither trial supports a recommendation about surgery beyond its own question. Compare only within each trial's own two arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

051015202530OS, median (months)20112024MARSMARS 2
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS, median, across 2 trials in this lane. MARS (2011): experimental 14.4 months vs control 19.5 months; MARS 2 (2024): experimental 19.3 months vs control 24.8 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
MARS2011OS19.5 months14.4 months
MARS 22024OS24.8 months19.3 months

Prophylactic radiotherapy to pleural procedure tracts: tract metastasis within 12 months

People with histocytologically proven pleural mesothelioma who had a large-bore pleural intervention in the previous 35 days · Any setting · endpoint: Procedure-tract metastasis incidence

Results are not comparable across trials.

This lane holds one randomised trial, SMART, so nothing is compared across trials. The plotted value is a rate, not a median: the intention-to-treat incidence of procedure-tract metastasis within 7 cm of the intervention site within 12 months of randomisation, for immediate versus deferred radiotherapy (the control arm received radiotherapy only if a tract metastasis appeared). The per-protocol analysis and any other analysis are separate results and are not plotted. The earlier randomised trials (Boutin 1995, Bydder 2004 and O'Rourke 2007) have no per-arm values to plot in these records, and their units, doses and designs differ, so they are not pooled with SMART. Compare only SMART's own two arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100Procedure-tract metastasis incidence (%)2016SMART
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
Procedure-tract metastasis incidence, rate, across 1 trial in this lane. SMART (2016): experimental 9 % vs control 16 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
SMART2016Procedure-tract metastasis incidence16 %9 %

Standard of care, then and now

How the usual sequence of treatment changed for one setting. Choose a point in time; each step links to the milestone behind it.

Unresectable pleural mesothelioma: systemic regimens with a regulatory action

Adults with unresectable malignant pleural mesothelioma · Metastatic

  1. Chemotherapy1Pemetrexed + cisplatin chemotherapyPemetrexed plus cisplatin versus cisplatin in pleural mesothelioma (2003)
  2. Immunotherapy2Nivolumab + ipilimumabCheckMate 743: nivolumab plus ipilimumab versus chemotherapy (2020)
  3. Immunotherapy3Pembrolizumab + pemetrexed + platinumIND.227 / KEYNOTE-483: pembrolizumab added to platinum-pemetrexed (2023)
2025. Based only on the regulatory actions recorded in the cited milestones. FDA, 17 September 2024: pembrolizumab with pemetrexed and platinum chemotherapy, first line. EU, 4 April 2025: non-epithelioid disease only.
Text version
  • 2004: Pemetrexed + cisplatin chemotherapy
  • 2021: Pemetrexed + cisplatin chemotherapy → Nivolumab + ipilimumab
  • 2025: Pemetrexed + cisplatin chemotherapy → Nivolumab + ipilimumab → Pembrolizumab + pemetrexed + platinum

Reversals and dead ends

Approaches that were abandoned or withdrawn, and trials whose results were later questioned, are part of the story.

  • 2011

    MARS: extrapleural pneumonectomy versus no EPP, a feasibility study

    MARS was a small randomised feasibility study in the UK. After chemotherapy, 50 people with pleural mesothelioma were assigned to a major operation (extrapleural pneumonectomy) followed by radiotherapy, or to no operation. Median survival was 14.4 months in the operation group and 19.5 months without it, from small numbers. The authors concluded that a larger trial was not feasible.

    Randomised feasibility study, 12 UK hospitals: after induction platinum chemotherapy, 50 of 112 registered patients were randomised to EPP then hemithorax radiotherapy (24) or no EPP (26). OS HR for EPP versus no EPP 1.90 (95% CI 0.92-3.93; exact p=0.082); adjusted HR 2.75 (1.21-6.26; p=0.016). Median survival 14.4 versus 19.5 months. Serious adverse events: ten with EPP, two without.

  • 2015

    Vorinostat, NGR-hTNF, nintedanib, anetumab: no superiority shown

    Four randomised trials tested other drugs in pleural mesothelioma. Vorinostat, NGR-hTNF and anetumab ravtansine were tested after chemotherapy; nintedanib was added to first-line chemotherapy. None met its main goal of improving survival (two trials) or progression-free survival (two trials) over placebo or the comparison treatment. An earlier, smaller nintedanib study had looked more favourable.

    VANTAGE-014 (vorinostat versus placebo, n=661): median OS 30.7 versus 27.1 weeks, HR 0.98 (0.83-1.17). NGR015 (NGR-hTNF versus placebo, each with investigator-choice therapy, n=400): OS 8.5 versus 8.0 months, HR 0.94 (0.75-1.18). LUME-Meso phase 3 (first line, n=458): PFS 6.8 versus 7.0 months, HR 1.01 (0.79-1.30). ARCS-M (anetumab ravtansine versus vinorelbine, n=248): PFS 4.3 versus 4.5 months, HR 1.22 (0.85-1.74). Not pooled.

  • 2016

    Prophylactic radiotherapy to procedure tracts, 1995 to SMART

    Mesothelioma can grow along the track of a biopsy or drain, so radiotherapy to these sites has been tried as prevention. A small 1995 trial of 40 people found tract metastases in 8 of 20 without radiotherapy and none in 20 with it. Three later randomised trials, the largest in 203 people, found no significant difference.

    SMART (UK, 22 hospitals): 203 randomised to immediate (102) or deferred (101) radiotherapy, 21 Gy in three fractions; procedure-tract metastasis within 7 cm at 12 months in 9 (9%) versus 16 (16%), OR 0.51 (95% CI 0.19-1.32; p=0.14). Boutin 1995: 40 patients, none of 20 versus 8 of 20 (40%). Bydder 2004 (58 sites) and O'Rourke 2007 (61 patients): no significant difference. Units, doses and designs differ; not pooled.

  • 2017

    DETERMINE: tremelimumab did not improve survival over placebo

    In 571 people whose mesothelioma had progressed after one or two earlier treatments, the immunotherapy drug tremelimumab was compared with placebo. Median survival was 7.7 months with tremelimumab and 7.3 months with placebo, not a significant difference. Severe side effects were more common with tremelimumab (65% against 48%).

    DETERMINE, randomised (2:1) double-blind placebo-controlled phase 2b: tremelimumab 10 mg/kg every 4 weeks for 7 doses, then every 12 weeks, in pleural or peritoneal mesothelioma after one or two systemic regimens (n=571). Primary endpoint OS (ITT): median 7.7 months (95% CI 6.8-8.9) versus 7.3 (5.9-8.7); HR 0.92 (0.76-1.12), p=0.41. Grade 3 or worse adverse events 65% versus 48%.

  • 2024

    MARS 2: extended pleurectomy decortication plus chemotherapy

    MARS 2 randomised 335 people in the UK with pleural mesothelioma judged operable. After two cycles of chemotherapy, about half were assigned to an extended operation (pleurectomy decortication) plus more chemotherapy and about half to chemotherapy alone. Median survival was shorter with surgery (19.3 versus 24.8 months), and serious side effects were more common.

    Phase 3 pragmatic open-label RCT, 26 UK hospitals; 335 randomised after two chemotherapy cycles (169 surgery plus chemotherapy, 166 chemotherapy alone); primary outcome OS, ITT. Median OS 19.3 (IQR 10.0-33.7) versus 24.8 months (12.6-37.4); restricted mean survival time difference at 2 years -1.9 months (95% CI -3.4 to -0.3; p=0.019). Grade 3 or worse serious adverse events (events, not participants) 318 versus 169 (IRR 3.6; 95% CI 2.3-5.5).

  • 2025

    BEAT-meso: atezolizumab added to bevacizumab and chemotherapy

    In a trial of 400 people with untreated pleural mesothelioma, adding the immunotherapy atezolizumab to bevacizumab and chemotherapy delayed progression by about a month and a half and raised median survival by about two months, but the survival difference was not statistically significant, so the main goal was missed. Non-epithelioid disease appeared to gain more.

    Open-label phase 3 (ETOP 13-18), 400 randomised 1:1: atezolizumab 1200 mg plus bevacizumab 15 mg/kg plus carboplatin-pemetrexed (ABC) versus bevacizumab plus carboplatin-pemetrexed (BC). Primary endpoint OS in all randomised patients not met: median 20.5 versus 18.1 months, HR 0.84 (95% CI 0.66-1.06), P=0.14. Median PFS 9.2 versus 7.6 months, HR 0.72 (95% CI 0.59-0.89), P=0.0021. OS HR 0.51 non-epithelioid, 1.01 epithelioid (interaction P=0.012).

Today and tomorrow

Standard of care today

Prevention
The Health and Safety Executive describes mesothelioma as strongly associated with asbestos exposure, attributes most male cases to past occupational exposure and gives a latency of typically at least 30 years. The American Cancer Society says that mesothelioma is much more common in men than in women, that the US rate rose from the 1970s to the 1990s and then levelled off and declined slightly (changes it says are probably related to changes in workplace exposures to asbestos), and that the rate is still rising in many countries where workplace exposure is more common. Cancer Research UK states that 94% of mesothelioma cases in the UK are preventable; the page read gives no definition of that figure.
Health and Safety Executive 2026American Cancer Society 2026Cancer Research UK 2026
Localised
Two randomised trials of radical surgery in pleural mesothelioma are recorded here, and in both abstracts median survival was shorter in the surgery group (in MARS the 95% confidence interval of the overall survival hazard ratio includes 1). MARS (a feasibility study; 50 people randomised, 24 to extrapleural pneumonectomy and 26 to none) reported a hazard ratio for overall survival of 1.90 (95% CI 0.92 to 3.93; exact p=0.082) and median survival of 14.4 versus 19.5 months; its authors wrote that these data, although limited, suggest that extrapleural pneumonectomy within trimodal therapy offers no benefit and possibly harms patients. MARS 2 (335 randomised, 86% epithelioid) reported median survival of 19.3 months with extended pleurectomy decortication plus chemotherapy and 24.8 months with chemotherapy alone, a difference in restricted mean survival time at 2 years of -1.9 months (95% CI -3.4 to -0.3; p=0.019), and more grade 3 or worse serious adverse events with surgery. The abstract of the Society of Thoracic Surgeons 2026 expert consensus says that surgery, if appropriate, should be part of a multimodal plan and that pleurectomy/decortication or extended pleurectomy/decortication is strongly favoured over extrapleural pneumonectomy. A phase 2 trial of neoadjuvant nivolumab (16 people) or nivolumab with ipilimumab (14) in resectable epithelioid or biphasic disease reported that it met its co-primary safety and feasibility endpoints, with survival outcomes exploratory. The ASCO guideline update (2025), the ESMO guideline (2022) and the NCCN guideline (version 3.2026) are cited by identifier only; their recommendation text was not opened. The registered phase 3 AtezoMeso trial of adjuvant atezolizumab is listed under trials to watch.
Treasure T, Lang-Lazdunski L, Waller D, Bliss JM, Tan C, Entwisle J, et al 2011Lim E, Waller D, Lau K, Steele J, Pope A, Ali C, et al 2024Velotta JB, Roden AC, Rice J, Simone CB, Upadhyay B, Sood P, et al 2026Reuss JE, Lee PK, Mehran RJ, Hu C, Ke S, Jamali A, et al 2025Kindler HL, Ismaila N, Bazhenova L, Chu Q, Churpek JE, Dagogo-Jack I, et al 2025Popat S, Baas P, Faivre-Finn C, Girard N, Nicholson AG, Nowak AK, et al 2022Stevenson J, Riely GJ, Wood DE, Aisner DL, Axtell AL, Bauman JR, et al 2026ClinicalTrials 2026
Metastatic

FDA letters read:

  • NDA 21-462 (pemetrexed
  • signed 4 February 2004) approves its use in the treatment of patients with malignant pleural mesothelioma whose disease is unresectable or who are otherwise not candidates for curative surgery (the letter does not mention cisplatin)
  • BLA 125554/S-089 (nivolumab) and BLA 125377/S-115 (ipilimumab), both signed 2 October 2020, add the combination for adults with unresectable disease (the ipilimumab letter says first-line)
  • and BLA 125514/S-161 (pembrolizumab with pemetrexed and platinum chemotherapy
  • signed 17 September 2024) adds first-line treatment of unresectable advanced or metastatic disease. None of the letters uses accelerated-approval wording. In the EU, the Alimta EPAR page gives an authorisation date of 20 September 2004 for use with cisplatin in chemotherapy-naive patients with unresectable disease
  • the procedural-steps documents list the nivolumab with ipilimumab extension (WS/1881
  • CHMP opinion 22 April 2021, Commission decision 1 June 2021) and pembrolizumab with pemetrexed and platinum chemotherapy limited to unresectable non-epithelioid disease (II/0154
  • opinion 27 February 2025, decision 4 April 2025). The abstracts of the randomised trials report longer overall survival with the experimental arm in CheckMate 743 (nivolumab plus ipilimumab
  • median 18.1 versus 14.1 months
  • hazard ratio 0.74, 96.6% CI 0.60 to 0.91) and IND.227 (pembrolizumab added to chemotherapy
  • 17.3 versus 16.1 months
  • hazard ratio 0.79, 95% CI 0.64 to 0.98)
  • in BEAT-meso, adding atezolizumab to bevacizumab and carboplatin-pemetrexed did not meet its primary overall survival endpoint (hazard ratio 0.84, 95% CI 0.66 to 1.06
  • p=0.14). The FDA Humanitarian Device Exemption database lists the NovoTTF-100L system (H180002
  • decision date 23 May 2019) for adults with unresectable, locally advanced or metastatic disease, used with pemetrexed and platinum-based chemotherapy: a device pathway, not a drug approval. Guideline recommendation text (ASCO 2025, ESMO 2022, NCCN 3.2026) was not opened
U 2004U 2020U 2020U 2024European Medicines Agency 2026European Medicines Agency 2021European Medicines Agency 2021European Medicines Agency 2025US Food and Drug Administration 2026Baas P, Scherpereel A, Nowak AK, Fujimoto N, Peters S, Tsao AS, et al 2021Chu Q, Perrone F, Greillier L, Tu W, Piccirillo MC, Grosso F, et al 2023Felip E, Popat S, Dafni U, Ribi K, Pope A, Cedres S, et al 2025Kindler HL, Ismaila N, Bazhenova L, Chu Q, Churpek JE, Dagogo-Jack I, et al 2025Popat S, Baas P, Faivre-Finn C, Girard N, Nicholson AG, Nowak AK, et al 2022Stevenson J, Riely GJ, Wood DE, Aisner DL, Axtell AL, Bauman JR, et al 2026
Relapsed / refractory

No relapsed-setting drug indication was found in the US drug-label database:

  • an openFDA search of indication text for 'mesothelioma' on 25 October 2026 returned 50 labels (pemetrexed products, Opdivo, Yervoy, Keytruda and Keytruda Qlex), and in each the mesothelioma sentence specifies initial or first-line treatment. EU authorisations other than those named above were not searched. In randomised trials in previously treated disease the abstracts report: CONFIRM (nivolumab versus placebo, after platinum-based chemotherapy, pleural or peritoneal, 332 randomised) median overall survival of 10.2 versus 6.9 months (adjusted hazard ratio 0.69, 95% CI 0.52 to 0.91
  • p=0.0090)
  • PROMISE-meso (pembrolizumab versus gemcitabine or vinorelbine, after platinum-based chemotherapy, 144 randomised) no difference detected in progression-free survival by blinded independent central review (hazard ratio 1.06, 95% CI 0.73 to 1.53) or in overall survival (hazard ratio 1.12, 95% CI 0.74 to 1.69)
  • DETERMINE (tremelimumab versus placebo, after one or two previous systemic treatments, pleural or peritoneal, phase 2b, 571 randomised) no difference in median overall survival (7.7 versus 7.3 months
  • hazard ratio 0.92, 95% CI 0.76 to 1.12
  • p=0.41). A registered phase 3 trial of VT3989 versus gemcitabine or vinorelbine in epithelioid pleural mesothelioma after platinum-based chemotherapy and immunotherapy (sTEADfast) is not yet recruiting
US Food and Drug Administration 2026Fennell DA, Ewings S, Ottensmeier C, Califano R, Hanna GG, Hill K, et al 2021Popat S, Curioni-Fontecedro A, Dafni U, Shah R, O'Brien M, Pope A, et al 2020Maio M, Scherpereel A, Calabrò L, Aerts J, Perez SC, Bearz A, et al 2017ClinicalTrials 2026
Any setting
Peritoneal mesothelioma is the less frequent site: the Health and Safety Executive describes most cases as pleural, and the abstract of the NCCN pleural mesothelioma guideline puts pleural disease at approximately 85% of mesotheliomas. The records read contain no randomised comparison of cytoreductive surgery with HIPEC against systemic therapy. A multi-institutional registry of 405 people treated with cytoreductive surgery and HIPEC reported median survival of 53 months and 3- and 5-year survival of 60% and 47%; 31% had grade 3 to 4 complications and 9 people (2%) died perioperatively, and the authors conclude that the data suggest prolonged survival in selected people. A multisociety consensus using a GRADE approach (2025) unanimously recommended cytoreductive surgery with HIPEC for a carefully selected subset, a strong recommendation despite very low certainty of evidence. A national consensus update of the 2018 Chicago guidelines (Cancer 2025) reports over 90% consensus in all six of its pathway blocks in the second voting round, describes limited and inconclusive data on the timing of systemic therapy and records unanimous support for considering clinical trial enrolment. Two registered randomised trials are listed under trials to watch.
Health and Safety Executive 2026Stevenson J, Riely GJ, Wood DE, Aisner DL, Axtell AL, Bauman JR, et al 2026Yan TD, Deraco M, Baratti D, Kusamura S, Elias D, Glehen O, et al 2009Kusamura S, Cinquini M, Morris D, Piso P, Kindler H, Brandl A, et al 2025Brown LM, Wilkins SG, Bansal VV, Su DG, Gomez-Mayorga J, Turaga KK, et al 2025ClinicalTrials 2026ClinicalTrials 2026

Trials to watch

  • eVOLVE-Meso NCT06097728

    In unresectable pleural mesothelioma, does volrustomig (MEDI5752) with carboplatin and pemetrexed differ from the investigator's choice of platinum plus pemetrexed or nivolumab plus ipilimumab in overall survival (the registry's primary outcome)? Registry record: phase 3, randomised, open-label; 863 enrolled (actual); sponsor AstraZeneca; the registry description gives nivolumab plus ipilimumab as the comparator for non-epithelioid histology. Registry status: active, not recruiting (status verified 2026-10); no results posted.

  • sTEADfast NCT07824986

    In advanced epithelioid pleural mesothelioma previously treated with platinum-based chemotherapy and immunotherapy, does VT3989 differ from the investigator's choice of gemcitabine or vinorelbine in overall survival (the registry's primary outcome)? Registry record: phase 3, randomised, open-label; 350 estimated; sponsor Vivace Therapeutics. Registry status: not yet recruiting (status verified 2026-09; estimated start 2026-12-01); no results posted.

  • AtezoMeso NCT04996017

    After pleurectomy/decortication without macroscopic residual disease, does adjuvant atezolizumab differ from placebo in disease-free survival (the registry's primary outcome)? Registry record: phase 3, randomised 2:1, placebo-controlled, described in the registry text as double-blind (the structured masking field says single); 162 estimated; sponsor Gruppo Oncologico Italiano di Ricerca Clinica. Registry status: recruiting (status last verified 2025-03, so it may be out of date); no results posted.

  • HIT-Meso NCT05655078

    In unilateral malignant pleural mesothelioma (N0 or N1, M0) when the person and the multidisciplinary team opt for active surveillance and deferral of systemic therapy until progression, does hemithoracic proton beam therapy differ from the standard-of-care watch-and-wait approach in progression-free survival and overall survival (the registry's primary outcomes)? Registry record: randomised 1:1, open-label; registered as phase 'NA' although the registry summary calls it a phase III randomised controlled trial; 148 estimated; sponsor University College, London. Registry status: recruiting (status verified 2026-03); no results posted.

  • Chemotherapy with or without immunotherapy in peritoneal mesothelioma NCT05001880

    In peritoneal mesothelioma, does atezolizumab added to carboplatin, pemetrexed and bevacizumab differ from carboplatin, pemetrexed and bevacizumab alone in response rate (the registry's primary outcome)? Registry record: phase 2, randomised, open-label; the official title names neoadjuvant or palliative chemotherapy; 66 estimated; sponsor National Cancer Institute. Registry status: recruiting (status verified 2026-09); no results posted.

  • PTC-guided therapy in peritoneal mesothelioma NCT07330271

    In epithelioid or biphasic peritoneal mesothelioma scheduled for cytoreductive surgery with HIPEC and adjuvant therapy, does treatment guided by a patient-derived tumour-like cell cluster (PTC) model differ from standard chemotherapy in overall survival (the registry's structured primary outcome; its plain-language summary speaks of living longer without the cancer getting worse)? Registry record: randomised, open-label, registered as phase 'NA'; 347 estimated; sponsor Beijing Tsinghua Chang Gung Hospital. Registry status: recruiting (status verified 2026-06); no results posted.

References

Show all 101 references

Loading references… Open the list as a page

Every figure was checked against its primary source to the extent possible. Items that could not be fully confirmed are marked “needs review”. Educational history, not medical advice.

Report an error

Links open the primary source. A flag marked ⚑ needs review means a value or claim has not yet been independently confirmed.

2024ReversalSurgery

landmarkPerioperative

MARS 2: extended pleurectomy decortication plus chemotherapy

MARS 2 randomised 335 people in the UK with pleural mesothelioma judged operable. After two cycles of chemotherapy, about half were assigned to an extended operation (pleurectomy decortication) plus more chemotherapy and about half to chemotherapy alone. Median survival was shorter with surgery (19.3 versus 24.8 months), and serious side effects were more common.

Phase 3 pragmatic open-label RCT, 26 UK hospitals; 335 randomised after two chemotherapy cycles (169 surgery plus chemotherapy, 166 chemotherapy alone); primary outcome OS, ITT. Median OS 19.3 (IQR 10.0-33.7) versus 24.8 months (12.6-37.4); restricted mean survival time difference at 2 years -1.9 months (95% CI -3.4 to -0.3; p=0.019). Grade 3 or worse serious adverse events (events, not participants) 318 versus 169 (IRR 3.6; 95% CI 2.3-5.5).

According to the MARS 2 authors, surgery to remove all visible tumour was widely offered to people with resectable pleural mesothelioma and recommended in American and European guidelines and international consensus statements. They describe the evidence behind it as case series, which are open to selection bias, rather than a randomised comparison.[1]

An earlier UK study, MARS, had randomised 50 people between extrapleural pneumonectomy and no operation; it was a feasibility study and did not test the lung-sparing operation.[2]

MARS 2 recruited from 26 hospitals between June 2015 and January 2021. After two cycles of chemotherapy, 335 people were randomised: 169 to pleurectomy decortication or extended pleurectomy decortication followed by more chemotherapy, and 166 to chemotherapy alone. Overall survival was the primary outcome.[1]

Median survival was 19.3 months with surgery and 24.8 months with chemotherapy alone, and the difference in restricted mean survival time at two years was -1.9 months (95% CI -3.4 to -0.3; p=0.019). Serious adverse events of grade 3 or worse numbered 318 with surgery and 169 without, an incidence rate ratio of 3.6 (95% CI 2.3 to 5.5).[1]

The authors interpreted this as worse survival to two years, and more serious adverse events, with extended pleurectomy decortication.[1] The trial answers one question: this operation added to chemotherapy, in people judged resectable at UK surgical centres. It does not test other operations or other settings, and its registry entry lists an estimated enrolment of 328, with status unknown, rather than the 335 randomised.[3]

Standard before

According to the MARS 2 paper, surgery to remove all visible tumour (pleurectomy decortication or extended pleurectomy decortication) was widely offered to people with resectable pleural mesothelioma and recommended in American and European guidelines and international consensus statements. The paper describes the evidence behind it as case series, not a randomised comparison.

Why it mattered

MARS 2 compared surgery plus chemotherapy directly with chemotherapy alone. The authors report worse survival to two years and more serious adverse events with extended pleurectomy decortication, and state that the operation had never been evaluated in a randomised trial. The finding applies to this comparison in people judged resectable, in an open-label pragmatic trial.

Caveats

  • Open-label, pragmatic trial of one comparison: pleurectomy decortication or extended pleurectomy decortication (139 of 157 operated participants had the extended form, 13 the standard form) plus chemotherapy versus chemotherapy alone. It does not address extrapleural pneumonectomy, other operations or disease that is not resectable.
  • Proportional hazards did not hold, so the primary comparison used restricted mean survival time to 2 years. A Cox model restricted to the first 42 months gave HR 1.28 (95% CI 1.02-1.60); the paper states that the 30 participants in follow-up beyond 42 months were too few to compare. Five-year survival was similar (14% versus 13%).
  • The registry record (NCT02040272) lists an estimated enrolment of 328 and status unknown; the paper's planned sample size was 328 and it randomised 335. The paper's figure is used here.
  • Fewer participants in the surgery group (66 of 169, 39%) than in the chemotherapy-alone group (93 of 166, 56%) completed six cycles of first-line chemotherapy, and fewer received immunotherapy or another treatment known to improve overall survival (37, 22%, versus 64, 39%); 13 of 169 assigned to surgery crossed over to chemotherapy alone and 2 of 166 the other way.
  • Most participants were men (87%) and had epithelioid disease (86%); the paper reports a subgroup analysis by cell type that is not reproduced here.