Advanced thymic epithelial tumours after neoadjuvant therapy: a narrative review of surgical challenges and strategies Authors: Nabih Berjaoui, Micayla Pather, Eleni Karapanagiotou, Gianluca Lucchese, Akshay Patel, Andrea Billè Published: 2026 ABSTRACT AbstractBackground and ObjectiveThymic epithelial tumours, including thymoma and thymic carcinoma, are rare anterior mediastinal neoplasms. Complete surgical resection remains the strongest prognostic determinant, yet locally advanced disease frequently involves critical mediastinal structures and requires multimodality treatment. Neoadjuvant therapy is commonly used to improve resectability, although it alters operative anatomy in ways that can complicate subsequent surgery. This narrative review offers an updated appraisal of the surgical challenges that arise after neoadjuvant therapy, and of contemporary strategies for complex post-induction resection.MethodsA targeted literature search was carried out in PubMed, MEDLINE, Embase, Scopus, Web of Science and Google Scholar for English-language publications from January 1990 through January 2026. Search terms combined thymic malignancies, induction and chemoradiotherapy strategies, surgical resection, vascular reconstruction, pleural disease and survival. Emphasis was placed on contemporary literature (2010 to 2026) to reflect advances in imaging, multimodality therapy and surgical reconstruction.Key Content and FindingsNeoadjuvant therapy produces dense fibrosis and obscures tissue planes, which complicates the dissection of critical structures. Imaging often underestimates stromal remodelling and residual viable tumour, so unexpected intraoperative difficulty is common. Vascular involvement is a major driver of operative complexity and frequently calls for resection and reconstruction, with contemporary graft patency of roughly 85% to 90% at five years in specialised series. In stage III disease, reported five-year survival after complete resection is approximately 70% to 90% for thymoma and 30% to 50% for thymic carcinoma. In stage IVa disease, cytoreductive surgery remains a treatment option when macroscopic complete resection is feasible, but the balance between aggressive procedures such as extrapleural pneumonectomy and more conservative pleurectomy/decortication must be weighed very carefully, since high-quality comparative evidence is lacking. Emerging data suggest that neoadjuvant chemoradiotherapy may yield higher R0 rates than chemotherapy alone, albeit at the cost of more pronounced fibrosis, although a matching survival benefit has not yet been proven. Molecular profiling (including GTF2I and TP53/KIT alterations) and AI-assisted imaging biomarkers are increasingly informing patient selection and surgical planning.ConclusionsSurgery after neoadjuvant therapy is feasible and safe in experienced centres, and in selected patients the oncologic outcomes appear to approach those of upfront resection with acceptable postoperative morbidity. Given the rarity of these tumours and the retrospective nature of most of the evidence, however, any comparison with upfront surgery should be interpreted with caution. Future progress will depend on international collaboration, prospective registries, validated molecular and imaging biomarkers, and standardisation of multimodality strategies.Keywords: Neoadjuvant therapy, pleural dissemination, surgical resection, thymic epithelial tumours, vascular reconstruction Introduction Thymic epithelial tumours, which include thymoma and thymic carcinoma, are the most common primary neoplasms of the anterior mediastinum in adults, although they remain rare and biologically heterogeneous (1,2). Clinical outcomes depend on histologic subtype and stage, but complete surgical resection is consistently the strongest prognostic determinant across disease categories (3-5). In locally advanced TNM stage II, III or IVa disease, tumours often extend beyond the thymic capsule to involve the lung, pericardium, diaphragm, pleura, chest wall and major vessels, which substantially increases operative complexity (6-8). Even in these settings, extended surgical approaches, including vascular reconstruction, extrapleural pneumonectomy (EPP), and pleurectomy/decortication (PD), can still achieve meaningful long-term survival when complete (R0) resection is possible (7,9,10). Neoadjuvant therapy has become an important component of multimodality management for borderline-resectable or initially unresectable thymic epithelial tumours. Platinum-based chemotherapy remains the most commonly used induction strategy, and retrospective studies report radiologic regression, downstaging and improved resectability in selected patients (11-15). Response rates are in the range of 50% to 70% in thymoma, while responses in thymic carcinoma tend to be lower but clinically meaningful (14,16). By reducing tumour burden and occasionally improving surgical planes, neoadjuvant therapy may permit more complete resection and, in selected patients, reduce the need for highly morbid procedures (11,12). Despite better resectability, the effect of neoadjuvant therapy on operative anatomy and on surgical complexity is still incompletely defined. Post-treatment fibrosis, distortion of tissue planes and adherence to major vascular structures have all been described, although the reported severity, clinical relevance and predictors vary between series (17-20). Much of the available literature is limited by small cohorts, heterogeneous induction regimens, variable surgical techniques and inconsistent radiologic response assessment (18,19,21). These limitations complicate any attempt to identify the patients most likely to benefit from induction therapy, to anticipate intraoperative risk, or to optimise surgical strategy after treatment. Uncertainty therefore persists regarding the balance between oncologic gains from tumour downstaging and technical challenges introduced by treatment-related tissue remodelling. A focused review of contemporary evidence can help clarify the anatomic and technical challenges of surgery after neoadjuvant therapy, define the relationship between induction treatment, resectability and postoperative outcomes, and place extended surgical procedures (including vascular reconstruction and pleural cytoreduction surgery) within modern multimodality management. The objective of this narrative review is to provide an updated appraisal of the surgical challenges associated with neoadjuvant therapy in thymic epithelial tumours. The review examines the benefits and limitations of induction strategies in improving resectability, discusses contemporary approaches to complex mediastinal, pulmonary, pleural and vascular resection after induction therapy, and highlights emerging evidence and future directions aimed at optimising multimodality management of locally advanced and recurrent thymic malignancies. We present this article in accordance with the Narrative Review reporting checklist (available at https://med.amegroups.com/article/view/10.21037/med-2026-1-0012/rc). Methods A targeted literature search was carried out in PubMed, MEDLINE, Embase, Scopus, Web of Science and Google Scholar for English-language publications from January 1990 through January 2026. Reference lists of included articles and of relevant reviews were screened manually to identify additional eligible studies. The PubMed search string used as the primary query was: (“thymoma”[MeSH] OR “thymic carcinoma”[tiab] OR “thymic epithelial tumour*”[tiab] OR “thymic epithelial tumor*”[tiab]) AND (“neoadjuvant therapy”[MeSH] OR “induction therapy”[tiab] OR “neoadjuvant”[tiab] OR “chemoradiotherapy”[tiab]) AND (“surgery”[MeSH] OR “resection”[tiab] OR “resectability”[tiab] OR “vascular reconstruction”[tiab] OR “pleurectomy”[tiab] OR “extrapleural pneumonectomy”[tiab]). Equivalent term combinations were adapted for Embase (Emtree), Scopus, Web of Science and Google Scholar. Eligible studies examined neoadjuvant or induction therapy for thymoma or thymic carcinoma and reported outcomes relevant to surgical management, including resectability, intraoperative findings, vascular or pleural resection, recurrence, or survival. Eligible designs included retrospective and prospective cohorts, phase II trials, multicentre registry analyses, institutional surgical series, and relevant narrative or systematic reviews. Case reports were included only when they described a technique, reconstruction or complication pattern not otherwise represented in the literature. Studies without any relevance to operative or oncologic outcomes were excluded. After duplicate removal, records were screened by title and abstract by two independent reviewers (N.B. and A.P.). A total of 214 records were screened, of which 131 were excluded because they did not meet the predefined inclusion criteria. Full-text review was performed for 83 articles, and 31 were excluded for non-operative focus, irrelevant outcomes or insufficient data. The final evidence base comprised 52 studies. Any disagreement between the two primary reviewers was resolved by discussion, and any remaining disagreement was adjudicated by a senior third author (A.B.). A PRISMA-style flow diagram summarising identification, screening, eligibility and inclusion is shown in Figure 1. PRISMA-style flow diagram of study identification, screening, eligibility and inclusion. Records identified through database searching (PubMed, MEDLINE, Embase, Scopus, Web of Science, Google Scholar): n=287. Additional records identified through manual screening of reference lists: n=12. Records after duplicate removal: n=214. Records screened by title and abstract: n=214. Records excluded on the basis of title/abstract: n=131. Full-text articles assessed for eligibility: n=83. Full-text articles excluded, with reasons (non-operative focus, irrelevant outcomes, insufficient data): n=31. Studies included in the final evidence base: n=52. Priority was given to contemporary literature published between 2010 and 2026, while earlier influential studies were kept when they remained clinically or methodologically important. Given the narrative design, no formal risk-of-bias assessment was undertaken; instead, the strengths and limitations of the evidence are discussed in context. A summary of the search strategy is given in Table 1. Table 1Search strategy summary for the narrative reviewItemSpecificationDate of searchLiterature search conducted through January 2026Databases and other sources searchedPubMed, MEDLINE, Embase, Scopus, Web of Science, Google Scholar, and manual screening of reference lists from key articles and guidelinesSearch terms usedControlled vocabulary (MeSH) and free-text terms combining thymic epithelial tumours, thymoma, thymic carcinoma, neoadjuvant or induction therapy, chemoradiotherapy, surgery, resection, resectability, vascular resection and reconstruction, pleurectomy, extrapleural pneumonectomy, R0 resection, and surgical outcomes. Full PubMed string provided in the Methods sectionTimeframePrimarily 1990–2026, with emphasis on contemporary series (2010–2026)Inclusion criteriaNarrative reviews, systematic reviews, retrospective and prospective clinical series, multicentre cohorts and consensus guidelines addressing surgical management and multimodality treatment of thymic epithelial tumoursExclusion criteriaNon-English publications, isolated case reports without broader clinical or technical relevance, and studies not addressing surgical or oncologic outcomesSelection processStudy selection by two independent reviewers based on predefined inclusion criteria; disagreements resolved by a senior third reviewer acting as adjudicatorMeSH, Medical Subject Headings. Search strategy summary for the narrative review MeSH, Medical Subject Headings. Results and discussion Impact of neoadjuvant therapy on operative anatomy Neoadjuvant chemotherapy produces anatomic and histopathologic changes that directly affect operative strategy in thymic epithelial tumours. Treatment-related fibrosis, stromal reaction and desmoplastic remodelling are the main contributors to increased surgical complexity after induction therapy (17-19,21). Histologic studies show dense collagen deposition and loss of peritumoral fat planes, which obscure anatomic boundaries and complicate the dissection of the pericardium, pulmonary hilum and adjacent soft tissues (21,22). Similar post-induction changes have been reported in lung cancer surgery and offer a useful conceptual parallel (23,24). One consequence of this remodelling is a progressive loss of natural tissue planes, which limits safe mobilisation of the superior vena cava, innominate vein, aorta, pulmonary artery, phrenic nerves and pericardium even in patients who have shown a radiologic response. Imaging after neoadjuvant therapy tends to underestimate operative challenges. Contrast-enhanced computed tomography (CT) and magnetic resonance imaging (MRI) studies show that partial radiologic regression can conceal residual microscopic tumour and dense fibrotic adherence at tumour-organ interfaces (12,25,26). Data from the Italian TYME database indicate that radiologic and clinical assessment frequently underestimate stromal alteration and residual viable tumour compared with pathologic evaluation (25). Several imaging tools appear to improve prediction of invasion or response: contrast-enhanced CT with multiplanar reformats and cine or black-blood MRI sequences help characterise vascular adherence; perfusion CT and diffusion-weighted MRI help distinguish viable tumour from post-treatment fibrosis; and quantitative PET metrics (SUVmax, metabolic tumour volume and total lesion glycolysis) correlate with progression-free survival (26-28). Despite these tools, the imaging-pathology discordance after induction therapy remains significant and supports cautious interpretation of post-induction imaging during operative planning. Induction therapy also alters vascular tissue characteristics. Several reports describe increased friability and vascularity after chemotherapy, with increased intraoperative blood loss (median of approximately 1,300 mL in complex resections) and higher transfusion requirements compared with upfront surgery (10,29), and meticulous haemostasis is required (10,30). The extent of post-induction anatomic change appears to predict the technical scope of resection. Patients with persistent tumour contact or adherence to major vascular structures after neoadjuvant therapy are significantly more likely to require vascular resection and reconstruction. These observations support detailed preoperative imaging, including 3D-reconstructed CT and MRI, to anticipate vascular involvement and plan complex reconstruction (10,31). Neoadjuvant therapy therefore reshapes the operative field in thymic epithelial tumours. Although it may improve oncologic resectability, it also introduces technical challenges that call for advanced surgical expertise. Centres with high case volumes and experience in complex mediastinal and vascular surgery are best positioned to manage these competing effects and to achieve good outcomes. Surgical management of vascular involvement after neoadjuvant therapy Vascular involvement is a major determinant of surgical complexity in thymic epithelial tumours and often governs the feasibility of complete oncologic resection. Because the thymus sits near the superior vena cava, the innominate veins, the azygos vein, the pericardium and the pulmonary arteries, even limited tumour growth can bring the tumour into contact with or encase critical mediastinal structures (6-8). Neoadjuvant therapy may reduce tumour volume, but induction-related desmoplastic fibrosis often obscures dissection planes and can increase technical difficulty despite radiologic regression. Thymic epithelial tumours typically invade vascular structures by direct extension rather than by haematogenous spread, which sets them apart from lung cancer and mesothelioma. Histopathologic studies show that thymomas often infiltrate the vascular adventitia without breaching the intima, in keeping with relatively indolent behaviour and favourable outcomes after superior vena cava reconstruction in selected cases. Thymic carcinoma, by contrast, and particularly its squamous subtypes, shows more aggressive vascular invasion, including deeper mural involvement and occasional intraluminal extension or tumour thrombus (32-36). This biological distinction influences patient selection: thymomas with vascular involvement often remain candidates for curative resection, whereas thymic carcinoma warrants heightened caution and more complex multimodality planning. Preoperative assessment of vascular involvement after induction therapy remains imperfect. On CT and MRI, commonly used indicators include circumferential tumour contact, distortion of the vessel contour and venous narrowing; however, treatment-induced fibrosis can mimic invasion, and residual viable tumour can persist despite radiologic response. Reported sensitivities for predicting vascular invasion range from 40% to 70% (25,26,29,37). Advanced MRI techniques, including black-blood and cine sequences, can improve diagnostic accuracy in selected cases, although their broader availability remains limited. Given this uncertainty, surgical planning increasingly emphasises readiness for vascular resection and reconstruction in patients with persistent vessel contact after neoadjuvant therapy, with the understanding that definitive assessment of resectability often happens intraoperatively. Indications for vascular resection include persistent tumour-vessel abutment, circumferential encasement beyond critical thresholds, suspected adventitial invasion, and dense fibrotic adherence that prevents safe mobilisation of the vessel. Involvement of the innominate-superior vena cava junction increases the likelihood of reconstruction. In those cases early adoption of vascular control is important: proximal and distal superior vena cava control, pericardial release to improve vessel mobility, early division of the left innominate vein when its junction is involved, and readiness for cardiopulmonary bypass in selected high-risk cases (pulmonary artery or aortic arch involvement, or femoral vessel isolation for emergency cannulation) (17,18,21). Across contemporary thymic surgical series, cardiopulmonary bypass is needed in a minority of cases, generally 5% to 15%. In the institutional series of Patel et al., bypass was required in 2 of 14 patients (14.3%), reflecting selective use in anatomically high-risk resections (10). Several series show that isolated left innominate vein resection can often proceed without reconstruction because collateral venous pathways are usually sufficient, whereas superior vena cava resection almost always requires reconstruction to prevent venous hypertension and upper-extremity oedema. Reconstructive options include prosthetic interposition grafts, patch angioplasty with synthetic or biologic materials, and more complex bilateral innominate or atrial bypass techniques. With modern surgical practice and appropriate antithrombotic therapy, thymic-specific series report durable graft patency, with five-year rates of roughly 85% to 90% (10,32,33,38). The major published vascular reconstruction series are summarised in Table 2. Table 2Outcomes of vascular resection and reconstruction in thymic epithelial tumoursStudyPatients (n)Vessels reconstructedReconstruction techniqueR0 rateCPBGraft thrombosisPerioperative mortality5-year survivalPatel et al. (10)14SVC, innominate veinsProsthetic interposition, patch angioplasty~70–80%2/14 (14.3%)Low0%~60–70%Bertolaccini et al. (32)14SVCPTFE grafts~75–85%Rare14%0%~50%Zhang et al. (33)22Major veinsProsthetic reconstruction~80%OccasionalLow0%~44%Mendogni et al. (7)65Mixed great vesselsMixed techniques50.8%Not routineLow6.2%~60–80% (thymoma)Maurizi et al. (38)27SVC and innominate veinsProsthetic grafts, patches~80%SelectiveLow~4%58.1%CPB, cardiopulmonary bypass; PTFE, polytetrafluoroethylene; SVC, superior vena cava. Outcomes of vascular resection and reconstruction in thymic epithelial tumours CPB, cardiopulmonary bypass; PTFE, polytetrafluoroethylene; SVC, superior vena cava. Choice of graft material is a distinct technical decision and deserves dedicated consideration, particularly in the post-induction setting, where chemotherapy- and radiotherapy-related changes may favour one material over another. Autologous vein (most commonly spiral saphenous or femoral vein) offers excellent biocompatibility and low infection risk but is usually limited to smaller-calibre reconstruction because of diameter mismatch with the superior vena cava. Autologous or bovine pericardium provides a pliable biologic patch for angioplasty or tubular grafting, handles well in irradiated tissue, and has acceptable long-term patency, although calcification and late shrinkage have occasionally been described. Polytetrafluoroethylene (PTFE) and Dacron prostheses remain the workhorse for tubular superior vena cava replacement because of their size range and mechanical durability; reported five-year patency with PTFE is 80% to 90% in contemporary thymic series (10,32,33,38). Induction-related inflammation may raise the threshold for using synthetic material in heavily treated or contaminated fields, where biologic grafts are usually preferred to reduce the risk of graft infection. Irrespective of graft choice, most series use postoperative low-molecular-weight heparin bridging to warfarin or a direct oral anticoagulant for three to six months, with subsequent decisions based on imaging surveillance (10,32,38). Histology-driven surgical strategy: thymoma versus thymic carcinoma Although thymoma and thymic carcinoma are both discussed under the umbrella of thymic epithelial tumours, their biological behaviour and response to induction therapy differ substantially, and these differences should shape surgical strategy. Thymoma tends to be indolent, spreads by direct extension, rarely involves nodal basins and responds well to platinum-based chemotherapy (50% to 70% radiologic response rates). In contrast, thymic carcinoma is aggressive, frequently involves intrathoracic lymph nodes, shows deeper vascular and pericardial invasion, and responds less reliably to chemotherapy (11,14,16). In the post-induction setting, these differences translate into a different operative strategy. In thymoma, oncologic clearance is usually achievable with en bloc resection of involved structures and relatively narrow margins, because microscopic extension beyond the radiologic tumour is uncommon; selective lymph node sampling is generally considered sufficient. In thymic carcinoma, wider margins are preferred when anatomically feasible, because microscopic extension is more frequent and the lymph node basin (anterior mediastinal, paratracheal, subcarinal and internal mammary) must be systematically addressed. A comparative overview of surgical strategy between the two entities is provided in Table 3. Table 3Comparative surgical strategy in the post-induction setting: thymoma versus thymic carcinomaDimensionThymomaThymic carcinomaBiological behaviourIndolent, local spread predominantAggressive, higher nodal and systemic spreadChemotherapy responseHigher response rate (50–70%) with platinum-based regimensLower and less reliable response; limited pathologic complete responseMargin intentEn bloc resection with relatively narrow margins usually sufficientWider margins preferred where anatomically feasibleLymphadenectomySelective anterior mediastinal samplingSystematic sampling of anterior mediastinal and deep intrathoracic stations (paratracheal, subcarinal, internal mammary)Vascular invasion patternAdventitial, intima usually sparedDeeper mural invasion; occasional intraluminal thrombusRole of EPPConsidered selectively for bulky pleural disease after PD is judged insufficientRarely justified; outcomes poor, high morbidityRole of neoadjuvant immunotherapyLimited; risk of severe immune-related toxicity (e.g., myocarditis)Emerging role (phase II data for perioperative pembrolizumab)EPP, extrapleural pneumonectomy; PD, pleurectomy/decortication. Comparative surgical strategy in the post-induction setting: thymoma versus thymic carcinoma EPP, extrapleural pneumonectomy; PD, pleurectomy/decortication. Lymph node management after neoadjuvant therapy Lymph node management has historically received limited attention in thymic surgery, but both the ITMIG and the European Society of Thoracic Surgeons (ESTS) now recommend at least systematic sampling of the anterior mediastinal (N1) stations and, for thymic carcinoma or suspected invasive disease, extension to the deep intrathoracic (N2) stations including paratracheal, subcarinal and internal mammary nodes (21). Nodal involvement is reported in 25% to 30% of thymic carcinoma, 10% to 15% of WHO B3 thymoma, and less than 5% of A, AB, B1 and B2 thymoma, which supports a histology-driven approach to the extent of lymphadenectomy (34,35). Neoadjuvant therapy adds a further layer of complexity to nodal staging. Induction can downstage involved nodes pathologically, can mimic reactive nodal enlargement on imaging, and can fibrose the perinodal fat so that frozen-section sampling becomes less reliable. For these reasons, selective intraoperative decisions based on nodal appearance are less trustworthy after induction therapy, and systematic sampling of at least the anterior mediastinal and, in thymic carcinoma, the deep intrathoracic stations is advisable even in patients who appear to have had a complete radiologic response. Neoadjuvant-induced fibrosis introduces additional operative challenges. Adventitial thickening may preclude circumferential vascular dissection, and fibrotic adhesions can tether the tumour to the superior vena cava or innominate veins, raising the risk of vascular injury. Scarring often extends beyond the radiologic tumour margin and therefore requires wider en bloc resection to achieve oncologic completeness. Tissue oedema and friability raise the risk of venous tears, and chemotherapy-related changes may obscure adjacent neural structures such as the phrenic nerves. Effective management of vascular involvement after neoadjuvant therapy rests on close multidisciplinary coordination between thoracic and cardiovascular surgeons, radiologists experienced in post-induction imaging, medical oncologists, anaesthesiologists managing haemodynamic changes during vascular clamping, and intensivists overseeing postoperative anticoagulation and graft surveillance (19,20). This collaboration is particularly important in the post-induction setting, where unexpected fibrosis or vascular adherence may require real-time modification of operative strategy. Long-term oncologic outcomes support aggressive vascular resection in carefully selected patients. Multiple series show that vascular resection does not independently compromise survival, and outcomes appear to improve when complete (R0) resection becomes feasible. Reported five-year survival ranges from 60% to 80% for thymoma (Mendogni et al. and Bertolaccini et al.) and from 30% to 50% for thymic carcinoma (Zhang et al. and multicentre series), depending on stage and histology (3,7,20,32,33). Recurrence at vascular resection margins is uncommon, and tumour biology (particularly WHO type B3 thymoma and thymic carcinoma) is a stronger predictor of recurrence than vascular involvement itself (32-36). Management of pleural dissemination (stage IVa) after neoadjuvant therapy Pleural dissemination, corresponding to stage IVa disease, is a distinct clinical subset of thymic epithelial tumours characterised by tumour implantation along the pleura, diaphragm and mediastinal surfaces. Approximately 10% to 20% of patients with thymoma present with pleural seeding at diagnosis, and the pleura remains the most common site of recurrence (7,39). Management in the post-neoadjuvant setting requires a multimodality approach that combines systemic therapy, surgical cytoreduction and, in selected cases, adjunctive regional treatments. Macroscopic complete resection, when achievable, remains the strongest determinant of long-term survival. Pleural dissemination reflects local implantation rather than haematogenous spread. Thymoma tends to follow an indolent pleural metastatic pattern, whereas thymic carcinoma shows more aggressive dissemination and a higher risk of recurrence after cytoreduction (37,40). After neoadjuvant therapy, pleural disease often shows a heterogeneous response, with regression of some nodules and persistence of occult microscopic disease, which complicates preoperative assessment. CT and positron emission tomography (PET) frequently underestimate pleural tumour burden after induction therapy, because metabolic suppression and fibrosis can mask residual disease, so definitive staging is largely intraoperative (26,29). Surgical cytoreduction has traditionally been central to the management of stage IVa disease, but the evidence base supporting it is limited. There are no randomised trials, and the retrospective series on which current practice rests are small, heterogeneous and subject to substantial selection bias. A therapeutic benefit of surgery on pleural metastases from thymic epithelial tumours has therefore never been demonstrated with methodological rigour. The principal options are PD, including extended PD, and EPP; radiotherapy and perfusion-based strategies represent additional non-operative or adjunctive tools. The choice between them should be individualised, with very careful attention to the risk-benefit balance. PD preserves lung parenchyma, has lower operative mortality and preserves pulmonary function, and is preferred for limited disease, particularly in thymoma (41,42). EPP enables en bloc resection of pleura, lung, diaphragm, pericardium and adjacent mediastinal tissue and has sometimes been used for bulky pleural involvement despite neoadjuvant therapy. Large series report five-year overall survival of approximately 60% to 87% in stage IVa thymoma when macroscopic complete resection is achieved after EPP, but EPP carries higher perioperative morbidity (30% to 50%) and operative mortality (3% to 10%) than PD (41,43,44). The indication for EPP is highly controversial, and the available evidence does not support presenting it as a standard or default approach; its use should be limited to selected patients in high-volume centres with full multimodality expertise. Comparative outcomes of PD and EPP are summarised in Table 4. Table 4Outcomes of surgical cytoreduction for pleural dissemination (stage IVa) in thymic epithelial tumoursStudyProcedure(s)Patients (n) Complete resection Operative mortality SurvivalFabre et al. (43)EPP (pleuropneumonectomy)17 Macroscopic complete resection in all resected patients No intraoperative deaths; 30-day mortality 17.6% (3/17) 5-year OS 60%; 10-year OS 30%Ishikawa et al. (45)Multimodality therapy; EPP in 4 of 1111 Not reported None reported 3 of 4 EPP patients alive without recurrenceOkuda et al. (46)Nationwide JART series; EPP vs limited resection136 Not reported Not reported Better prognosis with 10 or fewer pleural nodulesMoser et al. (44)EPP 40; total pleurectomy 23; local pleurectomy 88152 R0 77.3% Not reported 5-year OS 87.2%; 10-year OS 62.7%; thymic carcinoma 5-year OS 56%Chandarana et al. (41)EPP and extended pleurectomy/decorticationAs per reference (41) As per reference 41 As per reference 41 As per reference 41EPP, extrapleural pneumonectomy; JART, Japanese Association for Research on the Thymus; OS, overall survival; R0, complete microscopic resection. Outcomes of surgical cytoreduction for pleural dissemination (stage IVa) in thymic epithelial tumours EPP, extrapleural pneumonectomy; JART, Japanese Association for Research on the Thymus; OS, overall survival; R0, complete microscopic resection. When pleural disease is so extensive that upfront conservative surgery is unlikely to achieve clearance, the first question is whether surgery is indicated at all, rather than how best to proceed with it. In carefully selected cases, chosen on the basis of histology, timing of pleural relapse and the overall biological behaviour of the tumour, preoperative chemotherapy can be a reasonable option to reduce tumour burden and facilitate later cytoreduction. Reported response rates in pleural disease specifically are 40% to 60% in thymoma and lower in thymic carcinoma (11,14,16). Complete eradication of pleural disease remains uncommon, and although tumour regression occasionally permits de-escalation from EPP to PD, treatment-related fibrosis can also increase operative complexity (21,30,32). In indolent, slow-growing thymoma with favourable histology, in contrast, there is no convincing evidence that preoperative chemotherapy reduces the risk of pleural recurrence, and its use in this setting is not supported by the available data. Pleural dissemination frequently involves the diaphragm and pericardium, with diaphragmatic involvement reported in 30% to 50% of contemporary series (39,41); diaphragmatic and pericardial resection with reconstruction can be performed safely, and complete clearance of these structures is essential for a macroscopic complete resection. Neoadjuvant immunotherapy is an emerging investigational strategy, particularly for thymic carcinoma, where chemotherapy response remains limited. A phase II trial of perioperative pembrolizumab reported major pathologic responses in approximately 45% to 50% of patients and a pathologic complete response rate of 15% to 20%, with responses effectively confined to thymic carcinoma. Immune-related toxicity, including a rare but severe myocarditis signal in thymoma, underscores the need for histology-specific patient selection and careful safety monitoring (47). Hyperthermic intrathoracic chemotherapy (HITHOC) has been explored as a post-cytoreductive adjunct intended to eradicate residual microscopic disease after PD or EPP. Several series report improved local control and reduced pleural recurrence with heated cisplatin- or carboplatin-based perfusion, with acceptable morbidity in selected patients (42,48-51). From a surgical perspective, however, HITHOC adds meaningful technical burden to an already complex procedure. The heated chemotherapy is delivered into tissue that has often been primed by induction chemotherapy or radiotherapy, which increases friability and the risk of perioperative bleeding. The risk of bronchopleural fistula rises, particularly after EPP or after major bronchial or parenchymal resection, and prolonged air leak is common. Systemic absorption of cisplatin can contribute to acute kidney injury, which requires careful intraoperative fluid management and perioperative nephroprotection. There is also no standardisation of drug choice, dose, perfusion temperature or duration across centres, and randomised data comparing HITHOC with cytoreduction alone are lacking. HITHOC should therefore be considered only in experienced centres and only in patients whose anatomy, histology and performance status are judged to justify the additional technical risk. Patient selection remains critical. Patients with good performance status, unilateral thoracic disease, favourable response or stability after induction therapy, absence of extrathoracic metastasis and adequate cardiopulmonary reserve derive the greatest benefit from aggressive surgical management. In contrast, patients with refractory or rapidly progressive disease, particularly aggressive histology, usually benefit more from systemic therapy or clinical trial enrolment than from extensive cytoreduction. Overall, management of stage IVa pleural dissemination after neoadjuvant therapy rests on coordinated systemic therapy, precise surgical staging and, where appropriate, cytoreductive techniques chosen according to the individual risk-benefit balance rather than a single default algorithm. Both PD and EPP can be appropriate when macroscopic complete resection is achievable, and adjunctive strategies such as HITHOC may expand treatment options in selected cases. Long-term survival still depends on complete pleural clearance within a multidisciplinary framework, and any decision to pursue aggressive surgery must be weighed against the limited quality of the available evidence. Postoperative outcomes after neoadjuvant therapy In retrospective series, post-induction surgery in experienced centres carries morbidity and mortality comparable to upfront resection. Complication rates are in the range of 20% to 40%, in line with complex anterior mediastinal surgery; common events include pulmonary complications, atrial arrhythmias, phrenic nerve palsy, chylothorax and venous thrombosis (14,20,30). Park et al. and Lucchi et al. reported higher intraoperative blood loss after induction chemotherapy but no independent increase in postoperative complications (15,17). Operative mortality is 0% to 3% in thymoma and 2% to 5% in thymic carcinoma, rising after EPP (3% to 10%), major vascular reconstruction (up to 7%) and procedures requiring cardiopulmonary bypass (approximately 5% to 15%) (7,18,32,41,42). Contemporary vascular reconstruction series report early mortality below 3% and graft patency above 85% at five years (10,32,33). Complete resection remains the most consistent predictor of long-term oncologic outcome (1-5). Neoadjuvant therapy improves resectability in selected patients, with R0 rates of 70% to 90% in thymoma and 40% to 70% in thymic carcinoma (11,14,16,37). Na et al. reported a significantly higher R0 rate with neoadjuvant chemoradiotherapy (93.3%) than with chemotherapy alone (73.3%), together with better tumour regression grades (11). In stage III disease with vascular involvement, incorporation of vascular reconstruction yields R0 rates of 60% to 80% (7,10,33); in stage IVa disease, macroscopic complete resection rates of approximately 55% to 70% have been reported after PD or EPP (7,41,42). Favourable predictors include thymoma histology, limited vascular encasement, unilateral pleural disease and radiologic response or stability after induction therapy. Long-term survival varies by stage and histology but, in series achieving macroscopic complete resection, is broadly comparable to outcomes after upfront surgery. For thymoma, five-year overall survival is 60% to 90% and ten-year survival 40% to 70% after complete resection. Stage III disease is associated with five-year survival of 70% to 90% and ten-year survival of 50% to 70% (Mendogni et al., Bertolaccini et al., Zhang et al.); stage IVa disease with five-year survival of 60% to 80% and ten-year survival of 40% to 65% when macroscopic complete resection is achieved (Chandarana et al., Wang et al.) (3,7,15,18,33,41,42). Thymic carcinoma remains less favourable, with five-year overall survival of approximately 30% to 50% (37,40,41). Recurrence occurs in 20% to 35% of cases, most often in the pleura or mediastinum, and correlates more strongly with tumour biology (WHO B3 thymoma and thymic carcinoma) than with the use of induction therapy itself. Functional outcomes are generally acceptable. Lung-preserving approaches such as PD preserve postoperative pulmonary function better than EPP, and vascular reconstruction rarely causes durable functional impairment when graft patency is maintained. The conclusions above rest on retrospective cohorts with unavoidable selection bias and should be interpreted accordingly. Strengths and limitations of the current evidence The evidence base supporting neoadjuvant therapy followed by surgery in advanced thymic epithelial tumours derives largely from heterogeneous retrospective series and registries and therefore requires cautious interpretation. Strengths of the literature include progressive international collaboration and the harmonisation of staging and outcome reporting through the ITMIG-IASLC project, which has standardised definitions of stage, invasion, resectability and recurrence and now permits more reliable cross-study comparisons (2,4,21). Across independent series, complete resection consistently emerges as the strongest survival predictor regardless of stage, histology or induction use, and post-induction surgery remains feasible and safe in experienced centres, with morbidity and mortality comparable to upfront resection (3-5,11,13-16,20,30). Reproducible results have also been reported for vascular reconstruction and pleural cytoreduction in specialised institutions (7,10,33), and molecular profiling has begun to inform risk stratification, with GTF2I mutations associated with indolent thymoma and TP53 and KIT alterations linked to aggressive thymic carcinoma (34,35,52). Several methodological limitations nonetheless deserve emphasis. Almost all comparative data come from retrospective cohorts subject to substantial selection bias, since the choice between induction and upfront surgery is rarely random. Institutional expertise is a further confounder, as the centres reporting good post-induction outcomes are by definition high-volume units whose results may not generalise. Induction regimens vary widely, the definitions of vascular invasion, pleural dissemination, resection completeness and recurrence remain inconsistent between series, and radiologic-pathologic discordance after induction therapy further complicates selection and operative planning. Important gaps remain, particularly the underrepresentation of thymic carcinoma and the limited data on long-term functional and quality-of-life outcomes, including pulmonary function after pleural procedures and the durability of vascular grafts. Until higher-level evidence emerges, management will continue to depend on integrated evidence, expert judgement and institutional experience. Future directions Future progress will depend on biology-informed patient selection, more discriminating imaging, and surgical platforms suited to the post-induction field. Expanded genomic profiling through initiatives such as the THYMOGENE trial may sharpen prediction of induction response and guide histology-specific selection (52). On the imaging side, radiomic signatures, PET-derived volumetric metrics and AI-based morphology scoring have all been associated with histologic subtype, progression-free survival or mortality (27,28,53), and warrant prospective validation in dedicated post-induction cohorts. Three-dimensional vascular reconstruction from thin-slice CT, augmented reality overlay and intraoperative navigation offer a promising way to identify safe dissection planes in the fibrotic post-induction mediastinum. By projecting a patient-specific model of the tumour, great vessels and phrenic nerves onto the operative field, these tools can help anticipate vascular adherence, plan the sequence of resection and support intraoperative decisions about graft type and length. Prospective evaluation in thymic epithelial tumours remains limited. Finally, the field needs prospective trials. Adaptive designs that integrate chemotherapy, immunotherapy, targeted agents (54-56) and modern radiotherapy (intensity-modulated and proton therapy) are particularly suited to a rare disease, and standardisation of HITHOC technique and patient selection remains a priority for stage IVa care. Robotic platforms with restored force feedback may eventually extend minimally invasive surgery to selected post-induction cases, but current evidence does not support routine use. Despite real gains in resectability and survival, gaps remain in patient selection, treatment sequencing and durable disease control. Future progress will depend on the integration of biology-informed decision-making, advanced imaging, novel systemic therapies and refined surgical strategies. Surgical and perioperative innovation will continue to shape practice, including improvements in vascular graft technology and antithrombotic strategies, image-guided and navigation-assisted techniques, and ongoing standardisation of HITHOC for pleural disease. Optimisation of multimodality sequencing will also be central to individualising care, with clearer criteria for induction versus upfront surgery, risk-adapted adjuvant therapy, advanced radiotherapy techniques such as intensity-modulated and proton therapy, and adaptive strategies guided by early imaging response. Overall, integration of molecular profiling, advanced imaging, novel systemic therapies and surgical innovation within multidisciplinary care offers the strongest path toward improved oncologic and functional outcomes. Conclusions Neoadjuvant therapy has become an important option in borderline-resectable and locally advanced thymic epithelial tumours, improving resectability in selected patients and, in experienced centres, enabling complete resection in disease that would otherwise be inoperable. The accompanying technical challenges, fibrosis, distorted tissue planes and adherence to major mediastinal structures, are real but manageable, and contemporary series report oncologic outcomes broadly comparable to upfront resection when complete resection is achieved. Progress will hinge on better patient selection rather than more aggressive surgery. Biology-informed criteria, advanced imaging and prospective registries are needed to define who genuinely benefits from induction therapy, from extended vascular reconstruction or from pleural cytoreduction, so that complex surgery is offered only when it can plausibly improve long-term outcome. In stage IVa disease, both PD and EPP remain options when macroscopic complete resection is achievable, with survival determined primarily by completeness of resection rather than the operative approach; given the absence of randomised evidence, the indication for EPP should remain highly selective. The evidence base remains limited by disease rarity and the predominantly retrospective literature, and continued international collaboration and biomarker-informed selection will be needed to refine patient selection and treatment sequencing. Supplementary The article’s supplementary files as Acknowledgments This review builds on a lecture presented at the 15th ITMIG Annual Meeting (ITMIG 2025) by Andrea Billè. Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Footnotes Provenance and Peer Review: This article was commissioned by the Guest Editor (Malgorzata Szolkowska) for the series “The Series Dedicated to the 15th International Thymic Malignancy Interest Group Annual Meeting (ITMIG 2025)” published in Mediastinum. The article has undergone external peer review. Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://med.amegroups.com/article/view/10.21037/med-2026-1-0012/rc Funding: None. Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://med.amegroups.com/article/view/10.21037/med-2026-1-0012/coif). The series “The Series Dedicated to the 15th International Thymic Malignancy Interest Group Annual Meeting (ITMIG 2025)” was commissioned by the editorial office without any funding or sponsorship. The authors have no other conflicts of interest to declare. References Available at https://med.amegroups.com/article/view/10.21037/med-2026-1-0012/prf Supplementary Materials The article’s supplementary files as