Angiosarcoma is a highly malignant endothelial neoplasm that accounts for 1–2% of all soft tissue sarcomas in humans, and its marked biological aggressiveness constitutes a major clinical challenge. Early identification is particularly difficult in patients with a history of malignancies treated with radiotherapy, due to its histopathological heterogeneity and nonspecific clinical presentation, which often leads to confusion with other mesenchymal or epithelial neoplasms. We report the case of a 72-year-old woman with a history of cervical carcinoma treated with radiotherapy who presented with progressive pelvic pain, significant weight loss, and urinary symptoms. Initial imaging studies suggested peritoneal carcinomatosis, while multiple biopsies were negative for tumour recurrence. Given the discordance between morphological findings and initial immunohistochemical studies, a stepwise diagnostic approach using immunohistochemical panels was implemented. This approach confirmed peritoneal angiosarcoma following demonstration of CD31 positivity. This case highlights the importance of maintaining a high index of suspicion in patients with a history of pelvic irradiation, the utility of sequential diagnostic algorithms, the limitations of resource-limited settings, and the value of a multidisciplinary approach in the management of rare sarcomas.
El angiosarcoma es una neoplasia endotelial altamente maligna que representa entre el 1% y el 2% de todos los sarcomas de tejidos blandos y cuya agresividad biológica constituye un importante desafío clínico. Su identificación temprana es particularmente difícil en pacientes con antecedentes de neoplasias tratadas con radioterapia, debido a su heterogeneidad histopatológica y presentación clínica inespecífica, lo que favorece su confusión con otras neoplasias mesenquimales o epiteliales. Se presenta el caso de una paciente de 72 años con antecedentes de carcinoma cervical tratado con radioterapia, quien consultó por dolor pélvico progresivo, pérdida de peso significativa y síntomas urinarios. Los estudios de imagen sugirieron inicialmente carcinomatosis peritoneal, mientras que múltiples biopsias iniciales resultaron negativas para recidiva tumoral. Ante la discordancia entre los hallazgos morfológicos y los primeros estudios inmunohistoquímicos, se utilizó un abordaje diagnóstico escalonado mediante paneles inmunohistoquímicos, lo que permitió confirmar el diagnóstico de angiosarcoma peritoneal tras demostrarse positividad para CD31. Este caso resalta la importancia de mantener un alto índice de sospecha en pacientes con antecedente de radiación pélvica, la utilidad de algoritmos diagnósticos secuenciales, las limitaciones de los sistemas de salud con recursos restringidos y el valor del enfoque multidisciplinario en el manejo de sarcomas raros.
Angiosarcomas are malignant neoplasms of endothelial origin that, despite accounting for approximately 1–2% of soft tissue sarcomas, exhibit one of the highest levels of biological aggressiveness within this group of tumours.1–3 Their rarity, combined with a non-specific clinical presentation, makes this entity a genuine diagnostic challenge. In the abdominal setting, patients typically present with vague symptoms such as abdominal pain, distension, or ascites, often haemorrhagic, resulting in clinical conditions that mimic infectious or inflammatory conditions, including tuberculous peritonitis, thereby significantly delaying clinical suspicion.4
This initial diagnostic ambiguity is far from trivial: up to 85.7% of small-bowel angiosarcomas have been reported to be misdiagnosed during the preoperative period, leading to delays in the treatment of a malignancy characterised by early dissemination and a poor prognosis, with five-year survival rates of approximately 35%.2 In this context, the identification of predisposing factors, such as previous exposure to radiotherapy, is of particular importance, as it may help raise diagnostic suspicion and facilitate earlier recognition of the disease.4
From a histopathological perspective, peritoneal angiosarcoma with epithelioid morphology adds a further layer of complexity. Its architecture, consisting of solid sheets of large polygonal cells with little or no vascular channel formation, may mimic neoplasms of epithelial, melanocytic, or lymphoid origin.5,6 Consequently, diagnosis relies on an appropriate immunohistochemical strategy. Although classical markers such as CD34 may show limited sensitivity and can lose expression in poorly differentiated tumours, the co-expression of CD31 and ERG has emerged as a highly specific approach for confirming endothelial differentiation.7,8
Case reportA 72-year-old woman, a homemaker from an urban area, presented for evaluation with a four-month history of severe hypogastric pain (8/10 on the visual analogue scale), a sensation of a pelvic mass, intermittent vaginal bleeding, and an unintentional weight loss of approximately 10kg over the preceding six months. These symptoms were accompanied by urinary features, including dysuria and vesical tenesmus, as well as recent-onset dyspnoea.
Her medical history was notable for stage IIB cervical carcinoma diagnosed in 2002, which had been treated with external beam radiotherapy and chemotherapy, without brachytherapy. Given this history, tumour recurrence was initially considered; however, recent biopsies of the vaginal vault were negative for neoplastic infiltration. On admission, the patient was tachycardic (127bpm), had clinically evident ascites, and appeared cachectic (weight 48kg, height 155cm, body mass index 19.98kg/m2). Laboratory investigations revealed mild anaemia (haemoglobin 11.1g/dL) and slight leukocytosis (10.3×103/μL).
Initial imaging assessment by CT urography revealed a lesion located between the bladder and rectum, associated with bilateral hydronephrosis. Subsequently, magnetic resonance imaging of the abdomen and pelvis performed on 7 January 2025, twenty-three years after radiotherapy, demonstrated an ill-defined lesion located posterior to the pubic symphysis, anterior to the anterior bladder wall, and involving the bladder floor, with infiltration of these structures. The lesion measured 53mm×47mm×33mm and exhibited heterogeneous signal intensity on T2-weighted sequences and isointensity on T1-weighted images, with diffusion restriction and heterogeneous contrast enhancement following contrast administration. In addition, an approximately 12mm defect was observed in the anterior bladder wall, suggestive of perforation. The uterus was reported to be of normal size and morphology, with a 12mm cervix showing no pathological enhancement, and the ovaries were unremarkable. Left inguinal lymphadenopathy measuring up to 12mm, nodular thickening of the greater omentum with diffusion restriction, and a moderate amount of fluid within the abdominal and pelvic cavities were also identified. Furthermore, degenerative changes of the spine and multiple bilateral pulmonary and subpleural nodules measuring up to 12mm were reported.
Analysis of the ascitic fluid revealed a haemorrhagic appearance with the presence of hyperreactive mesothelial cells, without conclusive evidence of malignancy on the initial cytological examination. Histopathological assessment of the laparoscopic biopsy specimens demonstrated, on haematoxylin and eosin staining (Figs. 1 and 2), features suggestive of metastatic carcinoma. However, the initial immunohistochemical panel (CKAE1/E3, p40, p16) showed positivity only for Ki-67, with a high proliferative index, providing no discriminatory value regarding the histogenetic origin of the neoplasm. This discrepancy between the morphological findings and the immunohistochemical profile prompted further diagnostic investigation.
Histopathological examination with haematoxylin and eosin staining revealed an infiltrative malignant neoplasm composed of an irregular vasoformative proliferation with anastomosing vascular channels and solid areas. At higher magnification, marked hypercellularity was observed, consisting of large polygonal cells with abundant eosinophilic cytoplasm, pleomorphic nuclei, and a predominantly solid growth pattern, consistent with epithelioid morphology. These findings initially raised the differential diagnosis of metastatic carcinoma and other poorly differentiated epithelioid neoplasms, necessitating further immunohistochemical evaluation.
A stepwise immunohistochemical approach was undertaken. Desmin, SOX10, CD45, p63, chromogranin, and synaptophysin were negative, thereby excluding muscular, melanocytic, haematolymphoid, epithelial, and neuroendocrine differentiation. Subsequently, CD10 and CD68 were negative, whereas vimentin showed positive staining. In a further stage of the investigation, WT1 demonstrated focal nuclear reactivity; however, the absence of calretinin expression did not support mesothelial differentiation. Finally, S100, smooth muscle actin, h-caldesmon, MyoD1, and EMA were negative, while CD34 showed focal staining and CD31 was positive (Fig. 3). These findings confirmed endothelial differentiation and established the definitive diagnosis of peritoneal angiosarcoma.
This case highlights the critical importance of comprehensive, sequential immunohistochemical studies in the diagnosis of poorly differentiated neoplasms, particularly in cases with atypical clinical and histopathological presentations such as the present one.
DiscussionThe diagnosis of primary angiosarcomas arising in serosal membranes, such as the peritoneum or pleura, represents a significant challenge in diagnostic pathology.9 Owing to their extreme rarity and non-specific clinical presentation, typically characterised by massive ascites or peritoneal thickening, these neoplasms frequently mimic peritoneal carcinomatosis or infectious peritonitis from both clinical and radiological perspectives.4 In the absence of pathognomonic findings, definitive diagnosis relies exclusively on meticulous histopathological and immunohistochemical evaluation.10
Morphologically, the epithelioid variant of angiosarcoma is characterised by the proliferation of large atypical polygonal cells with abundant eosinophilic or amphophilic cytoplasm and prominent nucleoli, arranged in nests or solid sheets.2,3,6 This architectural pattern often lacks obvious vascular lumen formation, rendering it histologically indistinguishable from malignant neoplasms that are far more common within the peritoneum. Consequently, the principal differential diagnoses include metastatic carcinoma and epithelioid mesothelioma.3,9,11
The main diagnostic challenge in this setting lies in the potentially misleading immunophenotypic profile exhibited by these lesions. To resolve this overlap and confirm endothelial differentiation, the strategic selection of vascular markers is crucial. Historically, CD34 has been widely employed; however, its sensitivity in cytological specimens and biopsy samples has been shown to be suboptimal (approximately 60%), and its expression may be lost in poorly differentiated or dedifferentiated tumour areas.7 By contrast, CD31 has demonstrated high sensitivity and specificity for endothelial lineage.9 More recently, the incorporation of ERG, a nuclear transcription factor, has represented a significant advance in diagnostic practice.6 Co-expression of ERG and CD31 achieves a sensitivity approaching 100%; therefore, the combined use of a cytoplasmic or membranous marker (CD31) and a marker with well-defined nuclear staining (ERG) allows reliable confirmation of the diagnosis of epithelioid angiosarcoma while excluding an epithelial or mesothelial origin of the cellular proliferation.7,8 However, rare non-endothelial neoplasms, including a subset of CIC-rearranged sarcomas, may co-express ERG and CD31, representing a potential diagnostic pitfall; these markers should therefore be interpreted together with morphology and a broader immunohistochemical panel.12
Within the differential diagnosis, the possibility of metastatic amelanotic melanoma must be categorically excluded, as this entity may also exhibit epithelioid morphology and a solid growth pattern.6,10 Such exclusion is achieved by demonstrating negativity for specific melanocytic markers, including S-100 protein and HMB-45. In conclusion, the diagnostic approach to peritoneal epithelioid angiosarcoma requires a high index of suspicion and the application of a broad immunohistochemical panel that must include CD31 and ERG, in order to avoid misdiagnosis as carcinoma or mesothelioma, which could lead to inappropriate therapeutic decisions.
The implementation of advanced diagnostic protocols within public healthcare systems continues to face persistent structural barriers,10 particularly in middle-income countries such as Colombia. The availability of highly complex molecular testing, including next-generation sequencing (NGS) and fluorescence in situ hybridisation (FISH), is largely concentrated in referral centres located in major urban areas, thereby limiting timely access for patients managed in peripheral institutions. This technological centralisation creates a diagnostic bottleneck that restricts comprehensive molecular characterisation to only a proportion of cases, resulting in significant delays in diagnostic confirmation. In aggressive neoplasms, delays of several weeks are of particular clinical significance, given the rapid rate of tumour growth and the direct impact of diagnostic timing on prognosis and therapeutic decision-making.
In response to these limitations, innovative stepwise implementation strategies have emerged, including telepathology and collaborative hospital networks, which have shown potential to optimise scarce resources and reduce waiting times through shared diagnostic workflows.10 While progress is being made towards the broader adoption of molecular technologies, it remains essential to strengthen the fundamental pillars of diagnostic practice: mandatory multidisciplinary teams, standardised sampling protocols, and robust outcome-monitoring systems.
Peritoneal angiosarcoma represents a major diagnostic and therapeutic challenge in oncology. Its rarity, histopathological heterogeneity, and aggressive biological behaviour necessitate a multidisciplinary approach and the application of standardised diagnostic and management protocols. Although advances in immunohistochemistry (particularly CD31 and ERG) and molecular biology have improved diagnostic accuracy, their implementation in resource-limited healthcare systems, such as that of Colombia, remains constrained by significant barriers, including unequal access to technology, delays in diagnostic confirmation, and a shortage of appropriately trained specialists.
The reality of healthcare delivery in middle-income settings calls for innovative strategies that balance scientific rigour with practical feasibility. Initiatives such as the decentralisation of molecular diagnostic laboratories, the adoption of cost-effective diagnostic algorithms (with CD31 and Ki-67 as first-line markers), and the use of telepathology have emerged as pragmatic solutions to reduce existing disparities.10 Nevertheless, sustainable progress will require strengthening training in sarcoma pathology, optimising referral pathways between healthcare centres, and prioritising these rare neoplasms within national cancer control policies.
This case reaffirms that, when dealing with complex neoplasms, timely diagnosis should not be a limited privilege. The integration of international evidence with locally adapted approaches based on real-world epidemiological data is essential to improve clinical outcomes. Peritoneal angiosarcoma, rather than representing an isolated clinical challenge, symbolises the need to build resilient oncology systems in which diagnostic precision and equitable access to care progress hand in hand. Each advance in this direction will not only benefit patients affected by rare diseases, but will also contribute to raising the overall standards of oncology practice in resource-limited settings.
ConclusionPeritoneal angiosarcoma represents an extraordinarily complex diagnostic and therapeutic challenge in oncology. Its extreme rarity, histopathological heterogeneity, and aggressive clinical behaviour necessitate a multidisciplinary approach grounded in robust diagnostic criteria. Although advances in immunohistochemistry, particularly the combined use of CD31 and ERG, together with developments in molecular pathology, have improved diagnostic accuracy, their implementation remains uneven across different healthcare settings.10
In middle-income environments, this reality underscores the need for strategies that reconcile scientific rigour with practical feasibility. In this regard, the adoption of cost-effective diagnostic algorithms, such as the initial use of CD31 and proliferative markers including Ki-67, together with the progressive decentralisation of specialised techniques and the strengthening of telepathology services, represents a pragmatic approach to reducing disparities in access to accurate diagnosis.
Nevertheless, sustainable progress in the management of these neoplasms requires structural interventions, including strengthening training in sarcoma pathology, optimising referral and counter-referral pathways between centres of differing levels of complexity, and incorporating these entities into the priorities of public oncology policies. In this context, the integration of global evidence with locally adapted approaches supported by real-world epidemiological data emerges as a fundamental strategy for improving clinical outcomes. This case also highlights the diagnostic limitations inherent to resource-constrained settings. Beyond being an uncommon entity, peritoneal angiosarcoma underscores the need to develop resilient oncology systems in which diagnostic precision and equitable access converge as priority objectives. Progress in this direction will not only benefit patients affected by rare diseases, but will also contribute to the broader strengthening of oncological practice in resource-limited settings.
Author contributionsJonathan Augusto Rodríguez contributed to the acquisition of clinical and histopathological data, study planning, and preparation of the initial manuscript draft.
Nancy Johanna Mejía Vargas contributed to the analysis and interpretation of the histopathological findings, critically reviewed the intellectual content, and strengthened the academic framework of the manuscript.
Jorge Enrique Moreno Acuña participated in the diagnostic interpretation, critical revision of the manuscript, and validation of the findings.
The final version of the manuscript submitted for publication was reviewed and approved by all authors, thereby guaranteeing transparency and accountability in authorship.
Ethical considerationsThis case report was reviewed and approved by the Research Ethics Committee of Hospital Universitario del Valle “Evaristo García” E.S.E. (Project Code: RPC45), which determined that the study constituted a minimal-risk investigation in accordance with Resolution 008430 of 1993 issued by the Colombian Ministry of Health and Social Protection. Written informed consent was obtained from the patient for the use of clinical information and images for academic and scientific purposes.
Use of artificial intelligenceThe graphs, figures, and textual content presented in this study were prepared entirely by the authors without the use of artificial intelligence tools, machine learning systems, or automated text- or image-generation technologies. The analysis, interpretation of results, and preparation of the manuscript were conducted using conventional methods of scientific review, statistical analysis, and academic synthesis, thereby ensuring original authorship, methodological traceability, and compliance with current ethical and editorial standards.
Role of the sponsorsThere were no sponsors for this study. Accordingly, no external entity had any role in the study design, data collection, analysis or interpretation of the data, manuscript preparation, or the decision to submit the work for publication.
FundingThis study received no external funding. No grants, donations, or financial support were obtained from public or private institutions, pharmaceutical companies, technology innovation firms, or any other sponsoring organisations.
Conflicts of interestThe authors declare that they have no financial, academic, institutional, or personal conflicts of interest that could have influenced the preparation, interpretation, or publication of the results presented in this manuscript.
The authors wish to express their gratitude to the Department of Pathology of the Universidad del Valle for the academic and technical support provided throughout the development of this work, as well as for its assistance with the histopathological analysis, diagnostic discussions, and ongoing professional training, all of which contributed significantly to the completion and quality of this case report.
The followings are the supplementary data to this article:
The Spanish text of this article is available as supplementary material in Appendix 1.
