Hepatology (Baltimore, Md.) · 2026
Stroma-driven intertumoral and intratumoral heterogeneity of PD-L1 expression in cholangiocarcinoma: Implications for biomarker development and prognostic evaluation.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Saborowski Anna, Simon Peter, Zhao Jinbo, Marhenke Silke, Volk Valery, Reschke Josephine Sophie, Korosteleva Anastasiia, Yi Dawei, Reineke-Plaaß Tanja, Augustin Jeremy, Tran Doan Duy Hai, Ibarra-Arellano Miguel A, Schapiro Denis, Lemke Steffen, Krönke Nicole, Rapposelli Ilario Giovanni, Segatto Oreste, Bathon Melanie, Gerdes Christoph, Calderaro Julien, Kolch Walter, Zhernovkov Vadim, Feuerhake Friedrich, Vogel Arndt
Аннотация
BACKGROUND AND AIMS: Immune checkpoint inhibitors are now considered part of the standard of care in biliary tract cancer, based on a statistically significant but overall modest survival benefit in recent pivotal trials. Unlike in other gastrointestinal malignancies, PD-L1-based scores were not significantly associated with survival. Intrahepatic cholangiocarcinomas (iCCA) are morphologically heterogeneous tumors, with a subset of iCCA dominated by stroma-rich areas. We aimed at understanding the potential impact of the intertumoral and intratumoral heterogeneity on PD-L1 expression in iCCA.
METHODS: Bulk transcriptome analysis and multiplex immunohistochemistry were performed on 141 clinically annotated resected iCCA.
RESULTS: Molecular signatures are critically driven by the relative stroma content, with higher inflammatory gene expression scores in tumor microenvironment (TME)-rich compared with TME-poor tumors. In addition to this intertumoral heterogeneity, we observed a striking intratumoral heterogeneity reflected by an enrichment of PD-L1-expressing cells in stroma-dominant areas. The effect of intratumoral heterogeneity on PD-L1-based scores was confirmed by analyzing "virtual biopsies," that is, randomly selected adjacent tumor regions designed to mimic clinical biopsies, further highlighting the broad challenges associated with biomarker development using needle biopsies. By integrating clinical data, we provide evidence that the prognostic value of PD-L1-expressing tumor cells is overall modest and may depend not only on the magnitude of expression but also on their localization within the tumor.
CONCLUSIONS: Our data indicate that stroma content may be a surrogate for inflamed iCCA. Moreover, the intertumoral and intratumoral heterogeneity of iCCA poses a significant challenge to interpreting the PD-L1 signal. To establish tissue-based biomarkers in iCCA, a more granular annotation of their expression in different intratumoral compartments will be critical.
IMPACT AND IMPLICATIONS: Intrahepatic cholangiocarcinomas display a striking intertumoral and intratumoral heterogeneity, with broad variation in stroma content. We show that transcriptome-based inflammation-associated scores and signatures segregate with relative stroma content and that stroma areas in individual tumors are enriched in immune cells and PD-L1-positive cells. We argue that because of the extensive intratumor spatial heterogeneity of PD-L1 expression-the assessment of which is exacerbated by the limited amount of tissue routinely obtained in diagnostic biopsies in the locally advanced/metastatic setting-PD-L1-based metrics are likely to be bound to lack robustness as predictive biomarker of response to checkpoint inhibitors in iCCA and may partially account for the lack of correlation between PD-L1-based scores and clinical outcome in the pivotal phase 3 trials. The general implications of the above are: future attempts to identify tissue-based biomarkers should take spatial variation in the tumor microenvironment (TME) into account; TME-rich tumors are inflamed; and the predictive relevance of stroma content should be further investigated.
METHODS: Bulk transcriptome analysis and multiplex immunohistochemistry were performed on 141 clinically annotated resected iCCA.
RESULTS: Molecular signatures are critically driven by the relative stroma content, with higher inflammatory gene expression scores in tumor microenvironment (TME)-rich compared with TME-poor tumors. In addition to this intertumoral heterogeneity, we observed a striking intratumoral heterogeneity reflected by an enrichment of PD-L1-expressing cells in stroma-dominant areas. The effect of intratumoral heterogeneity on PD-L1-based scores was confirmed by analyzing "virtual biopsies," that is, randomly selected adjacent tumor regions designed to mimic clinical biopsies, further highlighting the broad challenges associated with biomarker development using needle biopsies. By integrating clinical data, we provide evidence that the prognostic value of PD-L1-expressing tumor cells is overall modest and may depend not only on the magnitude of expression but also on their localization within the tumor.
CONCLUSIONS: Our data indicate that stroma content may be a surrogate for inflamed iCCA. Moreover, the intertumoral and intratumoral heterogeneity of iCCA poses a significant challenge to interpreting the PD-L1 signal. To establish tissue-based biomarkers in iCCA, a more granular annotation of their expression in different intratumoral compartments will be critical.
IMPACT AND IMPLICATIONS: Intrahepatic cholangiocarcinomas display a striking intertumoral and intratumoral heterogeneity, with broad variation in stroma content. We show that transcriptome-based inflammation-associated scores and signatures segregate with relative stroma content and that stroma areas in individual tumors are enriched in immune cells and PD-L1-positive cells. We argue that because of the extensive intratumor spatial heterogeneity of PD-L1 expression-the assessment of which is exacerbated by the limited amount of tissue routinely obtained in diagnostic biopsies in the locally advanced/metastatic setting-PD-L1-based metrics are likely to be bound to lack robustness as predictive biomarker of response to checkpoint inhibitors in iCCA and may partially account for the lack of correlation between PD-L1-based scores and clinical outcome in the pivotal phase 3 trials. The general implications of the above are: future attempts to identify tissue-based biomarkers should take spatial variation in the tumor microenvironment (TME) into account; TME-rich tumors are inflamed; and the predictive relevance of stroma content should be further investigated.