Adverse Prognostic Impact of Transitional and Pleomorphic Patterns in Pleural Nonepithelioid Mesothelioma: Insights From Comprehensive Analysis and Reticulin Stain Open Access

A retrospective analysis was carried out on 132 cases of pleural mesotheliomas, comprising 87 biphasic and 45 sarcomatoid subtypes, consecutively enrolled between 1980 and 2021. These cases underwent biopsy during video-assisted thoracoscopy and received diagnoses at the Pathology Section of Bari University, Bari, Italy. Epidemiologic data, encompassing sex, age, asbestos exposure, and survival, were extracted from medical records. Preoperative staging was assigned from imaging reports (computed tomography scan or metabolic information). Patients were further stratified into 3 clinical groups: group 1 (stage I), group 2 (encompassing patients with clinical stages II and III because of similar survival times), and group 3 (stage IV).

Following histologic diagnosis, patients received either palliative treatment (typically pemetrexed-cisplatin) or adjuvant chemotherapy (usually gemcitabine-cisplatin), and in some instances, radiotherapy was administered. Exclusion criteria for the study included patients who underwent extensive pleurectomy or pleural decortication. Ethical committee approval (No. 5062/2016) was obtained, and written consent for diagnosis and research activities was provided by all patients upon hospital admission. Biopsy specimens were fixed in a 10% formalin solution, paraffin embedded, and stained with hematoxylin-eosin, periodic acid–Schiff, and periodic acid–Schiff diastase stains. The sarcomatoid morphology was classified into usual, pleomorphic, and transitional in both groups. Usual sarcomatoid morphology, including desmoplastic pattern, was composed of elongated/spindle cells (>2 times longer than wide) arranged in solid sheets or within a fibrous stroma. The pleomorphic morphology was composed of large, often discohesive, cells that varied in size and shape and had dense abundant eosinophilic cytoplasm and single or multiple irregular nuclei, which often contained 1 or several prominent nucleoli (Figure 1, A and B). The transitional morphologic aspect was diagnosed in cases with large cohesive epithelioid cells characterized by well-defined cell borders but lacking overtly sarcomatoid features (Figure 1, C). The progressive loss of cellular cohesion among epithelioid, transitional, and sarcomatoid forms is accompanied by increased deposition of extracellular matrix rich in collagen, composed of reticulin fibers. These fibers are easily highlighted with a silver impregnation technique, such as the Gomori reticulin stain. This method was performed in all histotypes, primarily with the aim of identifying the sarcomatoid pattern, considered present if it constituted more than 5% of the tumor (Figure 1, D).

An immunohistochemical panel, including cytokeratins 5/6, calretinin, Wilms tumor-1, HBME-1, carcinoembryonic antigen, thyroid transcription factor-1, Ki-67, and BRCA1-associated protein 1, was used. The Ki-67 proliferative index was calculated as the percentage of positive tumor cell nuclei from a total of 500 neoplastic cells at ×400 magnification (field area = 0.56 mm²). Various morphologic parameters, including the biphasic/sarcomatoid histotype, mitotic count (mitoses/2 mm²), necrosis, and the presence of usual, pleomorphic, and transitional patterns in sarcomatoid mesotheliomas and the sarcomatoid component of biphasic mesotheliomas (if present in >5% of cases), were evaluated on the hematoxylin-eosin–stained sections. In biphasic mesotheliomas, the epithelioid pattern was divided into 2 groups based on the prevalent pattern (solid versus nonsolid). Regarding the Ki-67 percentage, tumors were categorized into 2 groups based on the median value (Ki-67 ≤ 25% and Ki-67 > 25%).

Statistical analysis was conducted with SPSS software, version 18. Prognostic indices were assessed at the time of diagnosis and examined as continuous or categorical variables. Survival curves were constructed by using the Kaplan-Meier method, and differences were appraised with the log-rank test (P < .05 was deemed significant). A multivariate analysis was undertaken, incorporating all parameters, using the Cox model. The reproducibility t test was implemented to gauge the agreement among pathologists in the microscopic identification of the transitional pattern with and without Gomori staining.

The Table summarizes the clinical and pathologic characteristics of the case series. The study included a patient population with a mean age of 70.01 years, mostly comprising males. Average survival ranged from 1 to 39 months, with a mean of 9.6 months. Asbestos exposure was well documented in 99 of the 132 cases (75%). Patients were classified into different disease stages, with advanced-stage sarcomatoid mesothelioma associated with significantly worse survival (mean survival difference of 2.00 months, P = .02). No significant survival differences were observed among different stages of biphasic mesothelioma. Additionally, younger patients with better performance status had improved median survival (P = .02). Treatment approaches included adjuvant chemotherapy (mean survival, 12.6 months), combined chemo-radiotherapy (mean survival, 20.4 months), and palliative chemotherapy alone (mean survival, 3.3 months). Significant survival differences were noted between patients with biphasic and sarcomatoid mesothelioma. Among the 87 biphasic mesothelioma cases, 39 (44.8%) had a prevalent (present in more than or equal to 50% of the tumor) epithelioid component, whereas 48 (55.2%) had a prevalent sarcomatoid component. Furthermore, in the epithelioid component of biphasic mesothelioma, 72 (82.8%) showed a solid pattern. Survival was longer in biphasic (10.6 ± 0.9 months; 95% CI, 8.7–12.5 months) than sarcomatoid (7.7 ± 1.1 months; 95% CI, 5.5–9.8 months) mesotheliomas (P = .048). In biphasic mesothelioma, a prevalent epithelioid pattern resulted in longer survival than for a prevalent sarcomatoid pattern or sarcomatoid pure mesothelioma (P < .001).

Gomori staining revealed distinct patterns across different subtypes. In the epithelioid components, the reticulum stain revealed almost no deposition of collagen matrix between tumor cells. This aspect was more noticeable in the solid pattern areas (Figure 2, A and B). In the transitional pattern, silver impregnation revealed an intermediate pattern between epithelioid and sarcomatoid, with an increase in the reticulum surrounding small nests and sometimes individual cells (Figure 2, C and D). Pleomorphic mesothelioma displayed variable staining patterns, reflecting the heterogeneous cellular composition and architectural diversity of this subtype, including areas of thickened reticulin framework amid pleomorphic cells or complete absence of reticulin (Figure 2, E and F). In sarcomatoid mesothelioma, there was abundant extracellular collagen matrix, and the Gomori stain highlights a significantly thickened reticulum (Figure 2, G and H).

A transitional pattern was observed in 33 mesotheliomas (26 biphasic and 7 sarcomatoid), while pleomorphic morphology was found in 21 cases (9 biphasic and 12 sarcomatoid).

The immunohistochemical expression of BAP1 (using antibody clone C4, Santa Cruz Biotechnologies) was assessed in a subset of tumors where tissue availability allowed it, and the reaction was deemed suitable, based on an internal positive control. In particular, 25 cases of biphasic mesothelioma were evaluated, and nuclear expression loss was observed in 13 cases (52%), 3 of which exhibited a transitional pattern. Twelve cases of sarcomatoid mesothelioma were assessed, and expression loss was observed in 3 cases (25%), one of which had a transitional pattern. None of the analyzed cases of mesothelioma with a pleomorphic component showed loss of BAP1 expression. There was no statistically significant correlation between BAP1 expression in sarcomatoid and biphasic mesothelioma (P = .12).

No significant difference in survival was observed between transitional and pleomorphic groups (P = .55). In both biphasic and sarcomatoid mesothelioma groups, the survival analyses based on the prevalent morphologic pattern (usual versus transitional versus pleomorphic) revealed highly significant differences (P = .046) (Figure 3, A and B).

The mitotic count, stratified according to the criteria proposed by the WHO 2021 classification¹  for the epithelioid histotype, showed differences tending toward statistical significance (P = .06). Conversely, the proliferative activity, as assessed by Ki-67 percentage expression, displayed highly significant differences in survival, both for biphasic and sarcomatoid histotypes (P = .01). Multiple linear analysis, including Ki-67 percentage, chemotherapy, age, sex, stage, necrosis, mitotic count, and histotypes, indicated that only chemotherapy, a prevalent epithelioid morphology in biphasic mesothelioma, and Ki-67 were significant predictors of survival (P < .001, P = .001, and P = .05, respectively). To diagnose the transitional pattern, the intraobserver reproducibility test produced a κ-value of 0.23 to 0.37 without and κ of 0.62 to 0.77 with reticulin staining.

The discovery of the transitional pattern in sarcomatoid mesothelioma has engendered a wealth of implications that extend beyond the boundaries of conventional diagnostic and prognostic concerns.¹⁰  By delving deeper into the complexities and nuances of this enigmatic morphologic subtype, we can gain a more comprehensive understanding of its significance in the realm of mesothelioma.

The presence of the transitional pattern within sarcomatoid mesothelioma represents a facet of paramount clinical importance. Our extensive analysis revealed that its influence on survival may be context-dependent, with no significant effect when compared to pleomorphic morphology. However, in the specific subset of cases characterized by prevalent sarcomatoid morphology and the usual pattern, the transitional pattern was linked to substantially reduced median survival. This emphasizes the critical role of the transitional pattern in guiding clinical decisions and underscores its potential as a valuable prognostic marker.

The morphologic attributes of the transitional pattern are distinctive and warrant particular attention. Characterized by elongated, well-rounded cells forming sheetlike structures with abundant cytoplasm, this pattern occupies a unique position in the mesothelioma spectrum. Its atypical cellular architecture, a departure from the typical epithelioid configuration, and the absence of pronounced sarcomatous traits render it a subject of great intrigue for pathologists.

The unfavorable prognosis associated with the transitional pattern has been consistently reported in the literature. Studies, such as the one conducted by Churg et al,⁸  have drawn parallels between the transitional pattern and sarcomatoid mesothelioma, implying that this subtype carries a similarly adverse outlook. Dacic et al⁹  have further accentuated the importance of establishing standardized diagnostic criteria to enhance the precision of transitional pattern recognition and correlate it with patient survival. Our observations strengthen the prognostic weight of the transitional pattern and suggest that this morphologic aspect could be characterized by a more aggressive biological behavior than the usual sarcomatoid morphology.

The diagnostic interpretation of the transitional pattern does not come without its share of intricacies. Our investigation identified a lack of consensus among pathologists when attempting to identify this pattern, underscoring the pressing need for more robust diagnostic criteria. From a morphologic perspective, the transitional pattern represents an intermediate form between epithelioid and sarcomatoid mesothelioma, where the tumor cell morphology develops into sheets of elongated, plump cells with abundant cytoplasm that are starting to lose their epithelioid characteristics but are not overtly assuming a spindle-shaped form and lacking clear-cut sarcomatous features. Alongside these cytologic changes, the “transition” toward a purely sarcomatoid form is accompanied by remodeling of the extracellular matrix, characterized by an increased production of collagen, easily highlighted with a reticulin stain. While Gomori reticulin staining showed promise in improving diagnostic accuracy, the incorporation of artificial intelligence (AI) methods, as exemplified by the work of Galateau-Salle et al,⁷  stands out as a promising avenue for addressing the diagnostic challenges associated with the transitional pattern. Acknowledging the potential of AI in this context, particularly convolutional neural networks, holds promise for enhancing histopathologic image analysis to classify subtypes by extracting features such as texture and shape.^11,12  The integration of AI into diagnostic processes offers opportunities to augment traditional histopathologic evaluation methods. Through rigorous validation procedures ensuring accuracy across diverse datasets and continuous learning mechanisms refining models over time, AI presents a valuable tool for improving diagnostic accuracy and streamlining classification processes.^11,12  A machine-learning algorithm could be trained to classify tumors with a transitional pattern with high robustness, similar to what was achieved by Galateau-Salle et al.⁷  However, it is essential to recognize that AI should not replace the expertise of pathologists but rather complement their capabilities, fostering interdisciplinary collaboration and ultimately enhancing patient care outcomes.^11,12  Furthermore, it is important to emphasize that the Gomori stain represents a cost-effective method, easily performable in any laboratory, and that it can be integrated into an assessment by an AI algorithm. This has been done for hepatocellular carcinomas, where the recognition and quantification of the reticulum framework were entrusted to a supervised AI model and subsequently correlated with prognostic indicators.¹³ 

Transitional mesothelioma presents unique challenges in differential diagnosis owing to its morphologic overlap with other mesothelioma subtypes and various benign and malignant lesions. Histologically, transitional mesothelioma exhibits a combination of epithelioid and sarcomatoid features, making it particularly prone to confusion with biphasic mesothelioma. However, careful histopathologic evaluation, including assessment of architectural patterns, cytologic features, and immunohistochemical profiling, is essential for accurate subclassification. Differential diagnostic considerations include reactive mesothelial hyperplasia, biphasic mesothelioma, metastatic adenocarcinoma, and sarcomatoid carcinoma.^1,7  Reactive mesothelial hyperplasia, often associated with inflammatory conditions or asbestos exposure, may mimic transitional mesothelioma histologically, emphasizing the importance of clinical correlation and ancillary diagnostic techniques.^1,7  Additionally, metastatic adenocarcinomas from various primary sites can exhibit biphasic or transitional morphology, necessitating immunohistochemical analysis to discern the primary origin.⁷  Sarcomatoid carcinomas, such as spindle cell variants of lung adenocarcinoma or sarcomatoid renal cell carcinoma, may also resemble transitional mesothelioma, highlighting the importance of a comprehensive diagnostic approach.^1,7 

In our study, we acknowledge the importance of BAP1 as a prognostic and diagnostic marker in mesothelioma, recognizing its potential to provide valuable insights into subtype characterization and clinical outcomes. The literature supports the significance of BAP1 in mesothelioma subtypes, particularly highlighting its role in transitional and pleomorphic variants. Previous studies have suggested that retained BAP1 expression is commonly observed in these subtypes, consistent with our findings.¹⁴  In particular, a study unveiled a 44% incidence of BAP1 loss in transitional mesotheliomas, aligning with findings in biphasic mesotheliomas and exceeding those observed in sarcomatoid subtypes.⁷  Regarding the pleomorphic pattern, research on the molecular characterization of this histotype identified BAP1 gene alterations in approximately 40% of cases that also exhibited epithelioid and biphasic subtypes. Conversely, sarcomatoid mesotheliomas with a pleomorphic component showed no BAP1 alterations.¹⁵  However, the exact implications of BAP1 expression in transitional and pleomorphic mesotheliomas remain an area of ongoing research and debate. Some studies have proposed that retained BAP1 staining in these subtypes may indicate a favorable prognosis,¹⁶  while others have reported conflicting results or emphasized the need for further investigation.¹⁷  Moving forward, larger-scale and prospective studies are warranted to elucidate the precise relationship between BAP1 expression and morphologic subtypes, ultimately informing more comprehensive diagnostic and prognostic strategies in mesothelioma management.

The exploration of the transitional pattern in sarcomatoid mesothelioma should serve as a catalyst for future research endeavors. The unveiling of its genetic and molecular underpinnings, its potential classification as a distinct histotype, and the further development of AI-based diagnostic tools all represent promising directions for future investigations. Collaborative efforts among institutions and experts can provide fresh insights into this unique mesothelioma subtype and its impact on patient care.

This study offers insights into the diagnostic and prognostic implications of this unique morphologic subtype. However, several limitations must be acknowledged. The relatively modest sample size, retrospective design, and interobserver variability among pathologists are notable challenges. The absence of molecular analyses and the lack of specificity regarding treatment modalities further restrict the scope of the study. Despite these limitations, the research underscores the importance of standardizing diagnostic criteria and exploring the potential of AI in improving diagnostic accuracy. Multicenter studies with larger and more diverse datasets, as well as prospective designs, are crucial for a comprehensive understanding of the transitional pattern and its impact on patient outcomes.

While this study did not investigate the correlation between the transitional pattern and immunotherapy, exploring the potential connection between the transitional pattern in mesothelioma and immunotherapy opens an intriguing avenue of research. The unique cellular characteristics of the transitional pattern may profoundly influence immune responses and the effectiveness of immunotherapeutic strategies. The abundant cytoplasm and atypical cell morphology within this pattern could potentially create an immunologically distinct microenvironment, as previously described for mesothelioma with distinct molecular alteration.^18,19 

To delve into this uncharted territory, future research should prioritize unveiling the immunologic aspects of the transitional pattern. This includes the examination of immune cell infiltration and the identification of potential biomarkers that could serve as predictors of immunotherapy responsiveness in mesothelioma cases characterized by this pattern. Establishing clinical trials tailored to assessing the efficacy of immunotherapy in cases featuring the transitional pattern is of utmost importance to gain a deeper understanding of its true potential.

The presence of the transitional pattern in sarcomatoid mesothelioma holds compelling clinical and prognostic relevance. A comprehensive understanding of its impact, coupled with the development of standardized diagnostic criteria and advanced diagnostic methodologies, emerges as a pivotal pursuit for optimizing patient management within the complex domain of mesothelioma. This unique histologic subtype may also play a pivotal role in the realm of immunotherapy, offering an intriguing avenue for further research. The intricate nuances of the transitional pattern beckon researchers and clinicians to embark on a scientific odyssey, untangling its complexities and ultimately contributing to the advancement of our knowledge in the field of mesothelioma.

The authors would like to thank Gisella Agazzino, ScD, Adele Araci, ScD, and Angela Barile, ScD, for technical support.