Andreas Kleppe

• Julbø, Frida Marie Ihle; Henriksen, Audun Ljone; Pradhan, Manohar; Lindstrøm, Erika Kristina; Kostolomov, Ilyá & Oukrif, Dahmane [Show all 21 contributors for this article] (2026). Reliable classification of polyps based on artificial intelligence: a development and validation study. EClinicalMedicine. 93. doi: 10.1016/j.eclinm.2026.103826. Full text in Research Archive Show summary

  Background The shortage of pathologists presents a significant bottleneck in delivering timely and accurate diagnoses. To address this challenge, we developed the POLyp Artificial Intelligence-based RISk classifier (POLARIS), a prescreening tool designed to assist pathologists in handling the increasing volume of colorectal biopsies. Methods Using 15,079 whole-slide images (WSIs) from 2993 patients in the UK bowel cancer screening program between 2014 and 2018 and the histological diagnoses as ground truth, POLARIS was developed applying the open-source foundation model H-optimus-0, multiple instance learning, and a training scheme designed to provide robust models. Each WSI was categorised into one of five classes based on the diagnosis, reflecting increasing risk of malignancy, and regrouped into two broader categories for clinical interpretation. One category includes only samples not requiring pathological review, specifically those classified as normal tissue or tubular adenomas with low-grade dysplasia (LGD). The second category includes all other types of polyps, which are polyps recommended for review by a pathologist. A prespecified validation protocol defined how a specific model should be evaluated on a geographically external dataset comprising 10,842 WSIs from Cheltenham General Hospital between 2008 and 2019 and acquired using two scanners, Leica Aperio AT2 and Hamamatsu NanoZoomer XR. After validation, three experienced pathologists independently assessed the cases in which model prediction differed most from clinical diagnosis and, blinded to model output, reached a consensus. Findings In the external validation, POLARIS correctly identified 98.94% (95% CI 96.92–99.78%) of polyps with high-grade dysplasia (HGD) and adenocarcinoma as POLARIS-positive, and correctly classified 83.04% (81.66–84.36%) of normal and tubular adenomas with LGD as POLARIS-negative. The overall balanced accuracy was 86.65% for this prespecified primary analysis. In the tuning set, balanced accuracy was 87.64%, the sensitivity for polyps with HGD and adenocarcinoma was 100% (95% CI 97.14–100), and the specificity for normal and tubular adenomas with LGD was 81.89% (79.94–83.70). POLARIS achieved an area under the receiver operating characteristic curve (AUROC) in external validation of 0.9449 (95% CI 0.9384–0.9507) for distinguishing normal and tubular adenomas with LGD from polyps recommended for review by a pathologist and 0.9788 (0.9718–0.9844) for polyps with HGD and adenocarcinoma cases specifically. The model predicted the same class for both scanners in 97.93% (97.49–98.32%) of the cases. In the review of selected cases, the pathologists agreed with the model and not the clinical diagnosis in 92.5% of the cases. Heatmaps highlighting regions where the model indicates high-risk features closely correlated with areas annotated by expert pathologists as high-risk. Interpretation POLARIS has the potential to enhance diagnostic workflows in colorectal pathology by reliably identifying high-risk lesions and highlighting high-risk regions while substantially reducing the number of slides requiring pathologist review.

• Kildal, Wanja; Pradhan, Manohar; Kristensen, Gunnar S Balle; Vlatkovic, Ljiljana; Tobin, Kari Anne Risan & Hveem, Tarjei Sveinsgjerd [Show all 9 contributors for this article] (2026). Prognostic role of estrogen receptor, L1 cell adhesion molecule, and tumor grade in patients with endometrial carcinoma of no specific molecular profile. International Journal of Gynecological Cancer. ISSN 1048-891X. 36(5). doi: 10.1016/j.ijgc.2026.104669. Full text in Research Archive
• Skrede, Ole-Johan; Pradhan, Manohar; Isaksen, Maria; Hveem, Tarjei Sveinsgjerd; Vlatkovic, Ljiljana & Nesbakken, Arild [Show all 23 contributors for this article] (2026). Generalisation of automatic tumour segmentation in histopathological whole-slide images across multiple cancer types. NPJ precision oncology. 10. doi: 10.1038/s41698-026-01311-6. Full text in Research Archive Show summary

  Deep learning is expected to aid pathologists in tasks such as tumour segmentation. We developed a general tumour segmentation model for histopathological images and examined its performance in different cancer types. The model was developed using over 20,000 whole-slide images from over 4000 patients with colorectal, endometrial, lung, or prostate carcinoma. Performance was validated in pre-planned analyses on external cohorts with over 3000 patients across six cancer types. Exploratory analyses included over 1500 additional patients from The Cancer Genome Atlas. Average Dice coefficient was over 80% in all validation cohorts with en bloc resection specimens and in The Cancer Genome Atlas cohorts. No performance loss was observed when comparing the general model with single-cancer models specialised in cancer types from the development set. In conclusion, extensive and rigorous evaluations demonstrate that generic tumour segmentation by a single model is possible across cancer types, patient populations, sample preparations and slide scanners.

• Kalsnes, Joakim; Isaksen, Maria; Julbø, Frida Marie Ihle; Pradhan, Manohar; Kleppe, Andreas & Raedt, Sepp de [Show all 33 contributors for this article] (2026). Applicability of mitotic figure counting by deep learning: a development and pan‐cancer validation study. FEBS Open Bio. 16(7), p. 1259–1270. doi: 10.1002/2211-5463.70210. Full text in Research Archive Show summary

  Mitotic figure counting is an established measure of cell proliferation that is included in grading systems. We developed a deep learning method for mitotic figure counting and evaluated its prognostic impact in multiple external validation datasets. The deep learning method was trained in whole slide images of tissue sections stained with haematoxylin and eosin from a publicly available breast cancer dataset where mitotic figures have been annotated by expert pathologists. The final model was externally validated according to a protocol with predefined analyses of 14 571 patient samples from 13 patient cohorts from seven different cancer types. The predefined primary analysis was univariable Cox survival analysis of the number of mitotic figures detected per mm 2 . Automatic mitotic figure counting correlated well with known proliferation rates, and patients with more mitotic figures per mm 2 had significantly worse patient outcome in all the studied cancer types except colorectal cancer. This study demonstrates the practical potential of automated, deep learning‐based mitotic figure counting, both by automating pathology work and by suggesting expanded use in more cancer types, such as prostate cancer.

• Bøe, Vilde Schulerud; Kleppe, Andreas; Foersch, Sebastian; Wagner, Daniel-Christoph; Busund, Lill-Tove Rasmussen & Rivera, Adín Ramírez (2025). Low-Rank Adaptations for increased Generalization in Foundation Model features. Proceedings of Machine Learning Research (PMLR). doi: https:/openreview.net/pdf?id=0BJTRUVDf4. Full text in Research Archive Show summary

  For foundation models (FMs) to truly advance computational pathology, they must deliver consistent and reliable predictions under diverse, unseen test conditions. Without such robustness, clinical trust and widespread adoption remain out of reach. Although many FMs for histopathology now exist, they have to our knowledge not been systematically tested for robustness by external researchers on independent datasets. In this study, we evaluate the robustness of foundation model features on three separate histopathology datasets and find that their performance drops on external data. Our analysis also reveals that these models often encode dataset-specific information, limiting their generalizability. To address this issue, we train a Weight-Decomposed Low-Rank Adaptation (DoRA) with strong data augmentations to improve feature robustness. Our experiments show that models trained with this adapter exhibit fewer signs of dataset-specific information and may generate more robust features across domains. These results highlight the need for robustness testing and encourage incorporating robustness considerations into the development, training, and tuning of FMs for histopathology. The code for this work will be available at https://github.com/dsb-ifi/DoRA-for-FM-robustness

• Helgesen, Sander Elias Magnussen; Manet, Anthony; Cyll, Karolina; Tobin, Kari Anne Risan; Kjæreng, Marna Lill & Kostolomov, Ilya [Show all 15 contributors for this article] (2025). Reliable and Efficient Tissue Segmentation in Whole-Slide Images. Journal of machine learning research. ISSN 1532-4435. 2, p. 1–11. doi: https:/openreview.net/pdf?id=3gsNUjFi1q. Full text in Research Archive Show summary

  Whole-slide images in digital pathology often contain large regions of irrelevant background, making tissue segmentation an important preprocessing step in many applications. Traditional rule-based approaches to tissue segmentation often work quite well, but it is difficult to create general rules that cover all instances. We here apply an unmodified nnU-Net v2 training setup on downsampled whole-slide to develop and test an efficient and robust tissue segmentation model. The dataset contained nearly 30,000 images from slides with different tissue types, imaged using different scanners, and annotated using a semi-automatic workflow so that all annotations have been verified or made by human experts. This large, diverse dataset enables the training of a tissue segmentation model that generalizes well across different scanners and tissue types. We observed that our proposed model achieves similar or better accuracy than other deep learning models, while offering better robustness than simpler rule-based methods. The best compromise between inference speed and accuracy was observed using images at 10 \textmu m per pixel. Our approach can be used as an efficient and well-suited preprocessing step for computational pathology. Source code, Dockerfiles, and model weights are made publicly available at: \url{https://github.com/icgi/Reliable-and-Efficient-Tissue-Segmentation-in-Whole-Slide-Images}.

• Franken, Ingrid; Bakker, Marie-Christine; Elias, Sjoerd G.; Lacle, Miangela M.; Stathonikos, Nikolas & Rubio-Alarcon, Carmen [Show all 20 contributors for this article] (2025). Complementary value of a digital pathology biomarker to post-surgery circulating tumor DNA in risk stratification of stage III colon cancer patients receiving adjuvant chemotherapy. Journal of Clinical Oncology. ISSN 0732-183X. 43(16_suppl), p. 3604–3604. doi: 10.1200/jco.2025.43.16_suppl.3604. Full text in Research Archive Show summary

  3604 Background: The current standard of care for patients with stage III colon cancer (CC) is resection followed by adjuvant chemotherapy (ACT). About half of patients are cured by surgery and hence overtreated with ACT, whereas ~30% experience recurrence despite ACT. Several studies show that patients with no detectable circulating tumor DNA (ctDNA) after surgery are at a lower risk of recurrence (RR), although false negative ctDNA results remain a concern. Other studies show prognostic value of digital pathology biomarkers on resected CC tissue, like the Combined Analysis of Pathologists and Artificial Intelligence (CAPAI; Kleppe Lancet Oncol 2022). This study aimed to explore the potential added value of CAPAI to post-surgery ctDNA in risk stratification of patients with stage III CC receiving ACT. Methods: Patients were selected from the Prospective Dutch ColoRectal Cancer (PLCRC) cohort substudy PROVENC3 (Rubio-Alarcon AACR 2024), based on stage III CC treated with radical resection and adjuvant capecitabine or CAPOX. Post-surgery ctDNA status was determined using Labcorp Plasma Detect. From the resected tumor, a representative H&E slide was digitalized to generate a DoMore-v1-CE-CRC score, which was combined with the pT and pN stage and number of assessed lymph nodes for classification as CAPAI high-, intermediate- or low-risk. Three-year RR and Cox proportional hazard ratios (HR) were reported for ctDNA-based and CAPAI-based risk groups. Time to recurrence was compared between risk groups using the log-rank test. Results: Post-surgery ctDNA status and CAPAI risk classification were available for 163 patients. The 20 patients (12%) with detectable ctDNA had a higher recurrence risk (3-year RR 60% [32-77], HR 4.9 [2.5-9.6], p < 0.001) than patients with no detectable ctDNA (N = 143, 3-year RR 18% [11-25]). Within the subgroup with no detectable post-surgery ctDNA, 50 patients (35%) were classified as CAPAI high-risk. These CAPAI high-risk patients had a higher recurrence risk (3-year RR 35% [20-48], HR 4.2 [2.0-9.1], p < 0.001) than patients classified as CAPAI low/intermediate-risk, who were combined based on their observed similar RR (N = 93, 3-year RR 9% [3-15%]). Conclusions: In patients with stage III colon cancer treated with adjuvant CAP(OX), CAPAI risk classification has potential to further stratify RR in the subgroup with no detectable post-surgery ctDNA. These preliminary results suggest that CAPAI high-risk may help identify patients with false negative post-surgery ctDNA results. Over half of all patients had both no detectable ctDNA and were CAPAI low/intermediate-risk. Given their low RR in our preliminary results, future studies on larger patient cohorts should focus on the ability to combine biomarkers to select very low-risk patients and evaluate whether these patients can potentially be spared ACT.

• Kasius, J. C.; Kildal, Wanja; Vrede, S. W.; Askautrud, Hanne Arenberg; Pradhan, Manohar & Altena, A. M. van [Show all 29 contributors for this article] (2025). Surgical stage in the era of molecular profiling of endometrial cancer. European Journal of Cancer. ISSN 0959-8049. 233. doi: 10.1016/j.ejca.2025.116164. Full text in Research Archive
• Lindemann, Kristina Yvonne Kathe; Kildal, Wanja; Kleppe, Andreas; Tobin, Kari Anne Risan; Pradhan, Manohar & Mascialino, Barbara [Show all 11 contributors for this article] (2025). Real-world outcomes in molecular subgroups for patients with advanced or recurrent endometrial cancer treated with platinum-based chemotherapy. International Journal of Gynecological Cancer. ISSN 1048-891X. 35(5). doi: 10.1016/j.ijgc.2024.101618. Full text in Research Archive Show summary

  Objective There is scarce real-world evidence on patients with advanced/recurrent endometrial cancer treated with platinum-based chemotherapy. We assessed the oncological outcome in groups by molecular classification. Methods This retrospective cohort study included patients with advanced/recurrent endometrial cancer treated with platinum-based chemotherapy after hysterectomy at The Norwegian Radium Hospital, Oslo University Hospital, Norway, between January 2006 and December 2017. Patients were molecularly classified as pathogenic POLE mutated, mismatch repair deficient, p53 abnormal, or no specific molecular profile. Time-to-recurrence and cancer-specific survival were calculated. Results We identified 264 advanced-stage patients (stage III/IV) and 96 patients with recurrent disease. The molecular classification was prognostic for time-to-recurrence (p < .0001) and cancer-specific survival (p < .0001) in patients with advanced disease, but the outcome did not differ significantly by molecular groups in recurrent patients. In all molecular groups, patients with stage III disease had longer time-to-recurrence and cancer-specific survival compared to patients with stage IV disease. The worst outcome was observed in patients with p53 abnormal tumors with an HR of 1.57 (95% CI 1.07 to 2.30) for time-to-recurrence and HR of 1.78 (95% CI 1.19 to 2.65) for cancer-specific survival in stage III/IV disease and an HR of 1.45 (95% CI 0.83 to 2.52) for time-to-recurrence and HR of 1.60 (95% CI 0.99 to 2.68) for cancer-specific survival in patients with recurrent disease. The few patients with POLE mutated tumors had favorable outcomes despite the advanced/recurrent disease status. Conclusions Oncological outcomes differ by molecular groups, in particular among patients with advanced disease. Patients with p53 abnormal tumors have the worst outcome, while patients with POLE mutated tumors have favorable outcomes even with recurrent disease. Implementation of the addition of immunotherapy to chemotherapy is expected to lead to substantial improvement of outcome, particularly in patients with mismatch repair deficient advanced/recurrent disease. There is still a high unmet need in advanced/recurrent patients with p53 abnormal and no specific molecular profile tumors.

• Pradhan, Manohar; Kildal, Wanja; Vlatkovic, Ljiljana; Tobin, Kari Anne Risan; Lindemann, Kristina Yvonne Kathe & Kristensen, Gunnar S Balle [Show all 8 contributors for this article] (2025). Prognostic value of tertiary lymphoid structures in molecular subgroups of endometrial carcinoma. International Journal of Gynecological Cancer. ISSN 1048-891X. 35(7). doi: 10.1016/j.ijgc.2025.101915. Full text in Research Archive
• Kleppe, Andreas; Lindemann, Kristina Yvonne Kathe; Kildal, Wanja; Tobin, Kari Anne Risan; Pradhan, Manohar & Vlatkovic, Ljiljana [Show all 10 contributors for this article] (2024). Prognostic and therapeutic implication of molecular classification including L1CAM expression in high-risk endometrial cancer. Gynecologic Oncology. ISSN 0090-8258. 192, p. 80–88. doi: 10.1016/j.ygyno.2024.11.005. Full text in Research Archive Show summary

  Introduction. The role of molecular classification and L1CAM in high-risk endometrial cancer is uncertain. We aimed to determine the association of molecular profiling and L1CAM with patterns of relapse and survival.<p> <p>Material and methods. This retrospective cohort study included patients referred to Department for Gynecologic Oncology, Oslo University Hospital between January 1, 2006 and December 31, 2017. L1CAM expression and molecular profiling according to ProMisE was performed. Main outcome was time to recurrence (TTR) and cancer specific survival (CSS). <p>Results. Of 489 patients, 486 could be molecular classified. Thirty-seven (8 %) had POLE mutated tumors, 148 (30 %) had MMRd tumors, 189 (39 %) had p53 abnormal tumors, and 112 (23 %) had NSMP tumors. High L1CAM expression was observed in 256 (53 %), low in 227 (46 %) tumors (6 (1 %) missing). ProMisE was significant for TTR but not for CSS in multivariable analysis. L1CAM was significant in multivariable analysis for both TTR and CSS. In a multivariable model with ProMisE and L1CAM expression in the same multivariable model, ProMisE lost significance while L1CAM remained significant. Patients with POLE mutated tumors entailed an excellent prognosis while patients with p53 abnormal or L1CAM overexpressing tumors entailed a poor prognosis with a high frequency of distant recurrences. Patients with MMRd tumors, NSMP and p53 abnormal tumors with low L1CAM had an intermediate prognosis. <p>Conclusions. L1CAM is an additional adverse factor in the p53 abnormal and NSMP groups. These groups need special attention in studies intensifying adjuvant treatment.

• Lindemann, Kristina Yvonne Kathe; Kildal, Wanja; Kleppe, Andreas; Tobin, Kari Anne Risan; Pradhan, Manohar & Isaksen, Maria [Show all 10 contributors for this article] (2024). Impact of molecular profile on prognosis and relapse pattern in low and intermediate risk endometrial cancer. European Journal of Cancer. ISSN 0959-8049. 200. doi: 10.1016/j.ejca.2024.113584. Full text in Research Archive Show summary

  Introduction - The role of molecular classification in patients with low/intermediate risk endometrial cancer (EC) is uncertain. Higher precision in diagnostics will inform the unsettled debate on optimal adjuvant treatment. We aimed to determine the association of molecular profiling with patterns of relapse and survival.<p> <p>Material and methods - This retrospective cohort study included patients referred to The Norwegian Radium Hospital, Oslo University Hospital from 2006–2017. Patients with low/intermediate risk EC were molecularly classified as pathogenic polymerase epsilon (POLE)-mutated, mismatch repair deficient (MMRd), p53 abnormal, or no specific molecular profile (NSMP). The main outcomes were time to recurrence (TTR) and cancer-specific survival (CSS).<p> <p>Results - Of 626 patients, 610 could be molecularly classified. Fifty-seven patients (9%) had POLE-mutated tumors, 202 (33%) had MMRd tumors, 34 (6%) had p53 abnormal tumors and 317 (52%) had NSMP tumors. After median follow-up time of 8.9 years, there was a statistically significant difference in TTR and CSS by molecular groups. Patients with p53 abnormal tumors had poor prognosis, with 10 of the 12 patients with relapse presenting with para-aortic/distant metastases. Patients with POLE mutations had excellent prognosis. In the NSMP group, L1CAM expression was associated with shorter CSS but not TTR.<p> <p>Conclusions - The differences in outcome by molecular groups are driven by differences in relapse frequency and -patterns and demand a higher precision in diagnostics, also in patients with low/intermediate risk EC. Tailored adjuvant treatment strategies need to consider systemic treatment for patients with p53 abnormal tumors and de-escalated treatment for patients with POLE mutated tumors.

• Tomar, Dhananjay; Binder, Alexander & Kleppe, Andreas (2024). Are nuclear masks all you need for improved out-of-domain generalisation? A closer look at cancer classification in histopathology. Advances in Neural Information Processing Systems. ISSN 1049-5258. Full text in Research Archive Show summary

  Domain generalisation in computational histopathology is challenging because the images are substantially affected by differences among hospitals due to factors like f ixation and staining of tissue and imaging equipment. We hypothesise that focusing on nuclei can improve the out-of-domain (OOD) generalisation in cancer detection. Wepropose a simple approach to improve OOD generalisation for cancer detection by focusing on nuclear morphology and organisation, as these are domain-invariant features critical in cancer detection. Our approach integrates original images with nuclear segmentation masks during training, encouraging the model to prioritise nuclei and their spatial arrangement. Going beyond mere data augmentation, we introduce a regularisation technique that aligns the representations of masks and original images. We show, using multiple datasets, that our method improves OODgeneralisation and also leads to increased robustness to image corruptions and adversarial attacks. The source code is available at https://github.com/ undercutspiky/SFL/

• Cyll, Karolina Urszula; Haug, Erik Skaaheim; Pradhan, Manohar; Vlatkovic, Ljiljana; Carlsen, Birgitte & Löffeler, Sven [Show all 14 contributors for this article] (2024). DNA ploidy and PTEN as biomarkers for predicting aggressive disease in prostate cancer patients under active surveillance. British Journal of Cancer. ISSN 0007-0920. doi: 10.1038/s41416-024-02780-x. Full text in Research Archive Show summary

  Background: Current risk stratification tools for prostate cancer patients under active surveillance (AS) may inadequately identify those needing treatment. We investigated DNA ploidy and PTEN as potential biomarkers to predict aggressive disease in AS patients.<p><p> Methods: We assessed DNA ploidy by image cytometry and PTEN protein expression by immunohistochemistry in 3197 tumour-containing tissue blocks from 558 patients followed in AS at a Norwegian local hospital. The primary endpoint was treatment, with treatment failure (biochemical recurrence or initiation of salvage therapy) as the secondary endpoint. <p>Results: The combined DNA ploidy and PTEN (DPP) status at diagnosis was associated with treatment-free survival in univariable- and multivariable analysis, with a HR for DPP-aberrant vs. DPP-normal tumours of 2.12 (p < 0.0001) and 1.94 (p < 0.0001), respectively. Integration of DNA ploidy and PTEN status with the Cancer of the Prostate Risk Assessment (CAPRA) score improved risk stratification (c-index difference = 0.025; p = 0.0033). Among the treated patients, those with DPP-aberrant tumours exhibited a significantly higher likelihood of treatment failure (HR 2.01; p = 0.027). <p>Conclusions: DNA ploidy and PTEN could serve as additional biomarkers to identify AS patients at increased risk of developing aggressive disease, enabling earlier intervention for nearly 50% of the patients that will eventually receive treatment with current protocol.

• Lindemann, Kristina Yvonne Kathe; Kleppe, Andreas; Eyjólfsdóttir, Brynhildur; Danbolt, Svana Heimisdottir; Wang, Yun Yong & Heli-Haugestøl, Anne Gjertine [Show all 12 contributors for this article] (2023). Prospective evaluation of an enhanced recovery after surgery (ERAS) pathway in a Norwegian cohort of patients with suspected or advanced ovarian cancer. International Journal of Gynecological Cancer. ISSN 1048-891X. 33(8), p. 1279–1286. doi: 10.1136/ijgc-2023-004355. Full text in Research Archive Show summary

  Objective This prospective cohort study evaluated the introduction of an enhanced recovery after surgery (ERAS) pathway in a tertiary gynecologic oncology referral center. Compliance and clinical outcomes were studied in two separate surgical cohorts. Methods Patients undergoing laparotomy for suspected or verified advanced ovarian cancer at Oslo University Hospital were prospectively included in a pre- and post-implementation cohort. A priori, patients were stratified into: cohort 1, patients planned for surgery of advanced disease; and cohort 2, patients undergoing surgery for suspicious pelvic tumor. Baseline characteristics, adherence to the pathway, and clinical outcomes were assessed. Results Of the 439 included patients, 235 (54%) underwent surgery for advanced ovarian cancer in cohort 1 and 204 (46%) in cohort 2. In cohort 1, 53% of the patients underwent surgery with an intermediate/high Aletti complexity score. Post-ERAS, median fasting times for solids (13.1 hours post-ERAS vs 16.0 hours pre-ERAS, p&lt;0.001) and fluids (3.7 hours post-ERAS vs 11.0 hours pre-ERAS, p&lt;0.001) were significantly reduced. Peri-operative fluid management varied less and was reduced from median 15.8 mL/kg/hour (IQR 10.8–22.5) to 11.5 mL/kg/hour (IQR 9.0–15.4) (p&lt;0.001). In cohort 2 only there was a statistically significant reduction in length of stay (mean (SD) 4.3±1.5 post-ERAS vs 4.6±1.2 pre-ERAS, p=0.026). Despite stable readmission rates, there were significantly more serious complications reported in cohort 1 post-ERAS. Conclusions ERAS increased adherence to current standards in peri-operative management with significant reduction in fasting times for both solids and fluids, and peri-operative fluid administration. Length of stay was reduced in patients with suspicious pelvic tumor. Despite serious complications being common in patients with advanced disease undergoing debulking surgery, a causal relationship with the ERAS protocol could not be established. Implementing ERAS and continuous performance auditing are crucial to advancing peri-operative care of patients with ovarian cancer.

• Yang, Li; Yang, Jinlin; Kleppe, Andreas; Danielsen, Håvard Emil Greger & Kerr, David J. (2023). Personalizing adjuvant therapy for patients with colorectal cancer. Nature Reviews Clinical Oncology. ISSN 1759-4774. p. 67–79. doi: 10.1038/s41571-023-00834-2. Full text in Research Archive
• Lindemann, Kristina Yvonne Kathe; Danbolt, Svana Heimisdottir; Ramberg, Lene Kristine Marsby; Eyjólfsdóttir, Brynhildur; Wang, Yun & Heli-Haugestøl, Anne Gjertine [Show all 15 contributors for this article] (2023). Patient-reported nausea after implementation of an enhanced recovery after surgery protocol for gynae-oncology patients. International Journal of Gynecological Cancer. ISSN 1048-891X. 33(8), p. 1287–1294. doi: 10.1136/ijgc-2023-004356. Full text in Research Archive Show summary

  Objectives This study aimed to analyze the adherence to strategies to prevent post-operative nausea and vomiting after implementation of an enhanced recovery after surgery (ERAS) protocol for gynae-oncology patients. Patient-reported nausea before and after ERAS was also studied. Methods This prospective observational study included all patients undergoing laparotomy for a suspicious pelvic mass or confirmed advanced ovarian cancer before (pre-ERAS) and after the implementation of ERAS (post-ERAS) at Oslo University Hospital, Norway. Patients were a priori stratified according to the planned extent of surgery into two cohorts (Cohort 1: Surgery of advanced disease; Cohort 2: Surgery for a suspicious pelvic tumor). Clinical data including baseline characteristics and outcome data were prospectively collected. Results A total of 439 patients were included, 243 pre-ERAS and 196 post-ERAS. At baseline, 27% of the patients reported any grade of nausea. In the post-ERAS cohort, statistically significantly more patients received double post-operative nausea and vomiting prophylaxis (64% pre-ERAS vs 84% post-ERAS, p&lt;0.0001). There was no difference in the need for rescue medication (82% pre-ERAS vs 79% post-ERAS; p=0.17) and no statistically significant difference between pre- and post-ERAS or between the surgical cohorts in patient-reported nausea of any grade on day 2. Patients who reported none/mild nausea on day 2 had significantly less peri-operative fluid administered during surgery than those who reported moderate or severe nausea (median 12.5 mL/kg/hour vs 16.5 mL/kg/hour, p=0.045) but, in multivariable analysis, fluid management did not remain significantly associated with nausea. Conclusion Implementation of an ERAS protocol increased the adherence to post-operative nausea and vomiting prevention guidelines. Nausea, both before and after laparotomy, remains an unmet clinical need of gynae-oncology patients also in an ERAS program. Patient-reported outcome measures warrant further investigation in the evaluation of ERAS.

• Kleppe, Andreas; Skrede, Ole-Johan; Raedt, Sepp De; Hveem, Tarjei Sveinsgjerd; Askautrud, Hanne Arenberg & Jacobsen, Jørn E. [Show all 14 contributors for this article] (2022). A clinical decision support system optimising adjuvant chemotherapy for colorectal cancers by integrating deep learning and pathological staging markers: a development and validation study. The Lancet Oncology. ISSN 1470-2045. 23(9), p. 1221–1232. doi: 10.1016/S1470-2045(22)00391-6. Full text in Research Archive Show summary

  The DoMore-v1-CRC marker was recently developed using deep learning and conventional haematoxylin and eosin-stained tissue sections, and was observed to outperform established molecular and morphological markers of patient outcome after primary colorectal cancer resection. The aim of the present study was to develop a clinical decision support system based on DoMore-v1-CRC and pathological staging markers to facilitate individualised selection of adjuvant treatment. Methods We estimated cancer-specific survival in subgroups formed by pathological tumour stage (pT<4 or pT4), pathological nodal stage (pN0, pN1, or pN2), number of lymph nodes sampled (≤12 or >12) if not pN2, and DoMore-v1-CRC classification (good, uncertain, or poor prognosis) in 997 patients with stage II or III colorectal cancer considered to have no residual tumour (R0) from two community-based cohorts in Norway and the UK, and used these data to define three risk groups. An external cohort of 1075 patients with stage II or III R0 colorectal cancer from the QUASAR 2 trial was used for validation; these patients were treated with single-agent capecitabine. The proposed risk stratification system was evaluated using Cox regression analysis. We similarly evaluated a risk stratification system intended to reflect current guidelines and clinical practice. The primary outcome was cancer-specific survival. Findings The new risk stratification system provided a hazard ratio of 10·71 (95% CI 6·39–17·93; p<0·0001) for high-risk versus low-risk patients and 3·06 (1·73–5·42; p=0·0001) for intermediate versus low risk in the primary analysis of the validation cohort. Estimated 3-year cancer-specific survival was 97·2% (95% CI 95·1–98·4; n=445 [41%]) for the low-risk group, 94·8% (91·7–96·7; n=339 [32%]) for the intermediate-risk group, and 77·6% (72·1–82·1; n=291 [27%]) for the high-risk group. The guideline-based risk grouping was observed to be less prognostic and informative (the low-risk group comprised only 142 [13%] of the 1075 patients). Interpretation Integrating DoMore-v1-CRC and pathological staging markers provided a clinical decision support system that risk stratifies more accurately than its constituent elements, and identifies substantially more patients with stage II and III colorectal cancer with similarly good prognosis as the low-risk group in current guidelines. Avoiding adjuvant chemotherapy in these patients might be safe, and could reduce morbidity, mortality, and treatment costs.

• Sert, M Bilal; Kristensen, Gunnar S Balle; Kleppe, Andreas & Dørum, Anne (2021). Long-term oncological outcomes and recurrence patterns in early-stage cervical cancer treated with minimally invasive versus abdominal radical hysterectomy: The Norwegian Radium Hospital experience. Gynecologic Oncology. ISSN 0090-8258. p. 1–8. doi: 10.1016/j.ygyno.2021.05.028. Full text in Research Archive Show summary

  Objective To compare long-term oncological outcomes in early-stage cervical cancer (CC) patients treated with minimally invasive radical hysterectomy (MIRH) versus abdominal radical hysterectomy (ARH), with a focus on recurrence patterns, tumor sizes, and conization. Methods This single-institution, retrospective study consisted of stage IA1-IB1 (FIGO 2009) squamous cell carcinoma, adenocarcinoma, and adenosquamous carcinoma of the cervix, who underwent radical hysterectomy between 2000 and 2017. Results Of the 582 patients included, 353 (60.7%) underwent ARH, and 229 (39.3%) MIRH. The median follow-up was 14.4 years in the ARH group and 6.1 years in the MIRH group (p < 0.0001). Among the 96 stage IA patients, only 3 (3.1%) experienced recurrence. Among stage IB1 patients, the risk of recurrence, after adjusting for standard prognostic variables, was twofold higher in the MIRH group versus the ARH group (HR 2.73, 95% CI: 1.56–4.80), and the relative difference was similar in terms of risk of cancer-specific survival (CSS) (HR 3.04, 95% CI: 1.28–7.20) and overall survival (OS) (HR 2.35, 95% CI: 1.21–4.59). In stage IB1 ≤ 2 cm patients without conization MIRH was associated with reduced time to recurrence (TTR) (HR 4.00, 95% CI: 1.67–9.57), CSS (HR 3.71, 95% CI: 1.19–11.58) and OS (HR 3.02, 95% CI: 1.24–7.34). Intraperitoneal combined recurrences accounted for 12 of 30 (40.0%) recurrences in the MIRH group but were not identified after ARH (p = 0.0001). Conclusions MIRH was associated with reduced TTR, CSS and OS versus ARH in stage IB1 CC patients. The risk of peritoneal recurrence was high, even for tumors ≤2 cm without conization.

• Kleppe, Andreas; Skrede, Ole-Johan; Raedt, Sepp De; Liestøl, Knut; Kerr, David J. & Danielsen, Håvard Emil (2021). Designing deep learning studies in cancer diagnostics. Nature Reviews Cancer. ISSN 1474-175X. 21(3), p. 199–211. doi: 10.1038/s41568-020-00327-9. Full text in Research Archive Show summary

  The number of publications on deep learning for cancer diagnostics is rapidly increasing, and systems are frequently claimed to perform comparable with or better than clinicians. However, few systems have yet demonstrated real-world medical utility. In this Perspective, we discuss reasons for the moderate progress and describe remedies designed to facilitate transition to the clinic. Recent, presumably influential, deep learning studies in cancer diagnostics, of which the vast majority used images as input to the system, are evaluated to reveal the status of the field. By manipulating real data, we then exemplify that much and varied training data facilitate the generalizability of neural networks and thus the ability to use them clinically. To reduce the risk of biased performance estimation of deep learning systems, we advocate evaluation in external cohorts and strongly advise that the planned analyses, including a predefined primary analysis, are described in a protocol preferentially stored in an online repository. Recommended protocol items should be established for the field, and we present our suggestions.

• Nunes-Xavier, Caroline Elisabeth; Kildal, Wanja; Kleppe, Andreas; Danielsen, Håvard Emil Greger; Wæhre, Håkon & Llarena, Roberto [Show all 10 contributors for this article] (2021). Immune checkpoint B7-H3 protein expression is associated with poor outcome and androgen receptor status in prostate cancer. The Prostate. ISSN 0270-4137. doi: 10.1002/pros.24180. Full text in Research Archive Show summary

  Background Novel immune checkpoint-based immunotherapies may benefit specific groups of prostate cancer patients who are resistant to other treatments. Methods We analyzed by immunohistochemistry the expression of B7-H3, PD-L1/B7-H1, and androgen receptor (AR) in tissue samples from 120 prostate adenocarcinoma patients treated with radical prostatectomy in Spain, and from 206 prostate adenocarcinoma patients treated with radical prostatectomy in Norway. Results B7-H3 expression correlated positively with AR expression and was associated with biochemical recurrence in the Spanish cohort, but PD-L1 expression correlated with neither of them. Findings for B7-H3 were validated in the Norwegian cohort, where B7-H3 expression correlated positively with Gleason grade, surgical margins, seminal vesicle invasion, and CAPRA-S risk group, and was associated with clinical recurrence. High B7-H3 expression in the Norwegian cohort was also consistent with positive AR expression. Conclusion These results suggest distinct clinical relevance of the two immune checkpoint proteins PD-L1 and B7-H3 in prostate cancer. Our findings highlight B7-H3 as an actionable novel immune checkpoint protein in prostate cancer.

• Cyll, Karolina Urszula; Kleppe, Andreas; Kalsnes, Joakim; Vlatkovic, Ljiljana; Pradhan, Manohar & Kildal, Wanja [Show all 14 contributors for this article] (2021). PTEN and DNA Ploidy Status by Machine Learning in Prostate Cancer. Cancers. 13(17). doi: 10.3390/cancers13174291. Full text in Research Archive Show summary

  Machine learning (ML) is expected to improve biomarker assessment. Using convolution neural networks, we developed a fully-automated method for assessing PTEN protein status in immunohistochemically-stained slides using a radical prostatectomy (RP) cohort (n = 253). It was validated according to a predefined protocol in an independent RP cohort (n = 259), alone and by measuring its prognostic value in combination with DNA ploidy status determined by ML-based image cytometry. In the primary analysis, automatically assessed dichotomized PTEN status was associated with time to biochemical recurrence (TTBCR) (hazard ratio (HR) = 3.32, 95% CI 2.05 to 5.38). Patients with both non-diploid tumors and PTEN-low had an HR of 4.63 (95% CI 2.50 to 8.57), while patients with one of these characteristics had an HR of 1.94 (95% CI 1.15 to 3.30), compared to patients with diploid tumors and PTEN-high, in univariable analysis of TTBCR in the validation cohort. Automatic PTEN scoring was strongly predictive of the PTEN status assessed by human experts (area under the curve 0.987 (95% CI 0.968 to 0.994)). This suggests that PTEN status can be accurately assessed using ML, and that the combined marker of automatically assessed PTEN and DNA ploidy status may provide an objective supplement to the existing risk stratification factors in prostate cancer.

• Skrede, Ole-Johan; Raedt, Sepp De; Kleppe, Andreas; Hveem, Tarjei Sveinsgjerd; Liestøl, Knut & Maddison, John [Show all 20 contributors for this article] (2020). Deep learning for prediction of colorectal cancer outcome: a discovery and validation study. The Lancet. ISSN 0140-6736. 395:10221, p. 350–360. doi: 10.1016/S0140-6736(19)32998-8. Full text in Research Archive Show summary

  Background Improved markers of prognosis are needed to stratify patients with early-stage colorectal cancer to refine selection of adjuvant therapy. The aim of the present study was to develop a biomarker of patient outcome after primary colorectal cancer resection by directly analysing scanned conventional haematoxylin and eosin stained sections using deep learning. Methods More than 12 000 000 image tiles from patients with a distinctly good or poor disease outcome from four cohorts were used to train a total of ten convolutional neural networks, purpose-built for classifying supersized heterogeneous images. A prognostic biomarker integrating the ten networks was determined using patients with a non-distinct outcome. The marker was tested on 920 patients with slides prepared in the UK, and then independently validated according to a predefined protocol in 1122 patients treated with single-agent capecitabine using slides prepared in Norway. All cohorts included only patients with resectable tumours, and a formalin-fixed, paraffin-embedded tumour tissue block available for analysis. The primary outcome was cancer-specific survival. Findings 828 patients from four cohorts had a distinct outcome and were used as a training cohort to obtain clear ground truth. 1645 patients had a non-distinct outcome and were used for tuning. The biomarker provided a hazard ratio for poor versus good prognosis of 3·84 (95% CI 2·72–5·43; p<0·0001) in the primary analysis of the validation cohort, and 3·04 (2·07–4·47; p<0·0001) after adjusting for established prognostic markers significant in univariable analyses of the same cohort, which were pN stage, pT stage, lymphatic invasion, and venous vascular invasion. Interpretation A clinically useful prognostic marker was developed using deep learning allied to digital scanning of conventional haematoxylin and eosin stained tumour tissue sections. The assay has been extensively evaluated in large, independent patient populations, correlates with and outperforms established molecular and morphological prognostic markers, and gives consistent results across tumour and nodal stage. The biomarker stratified stage II and III patients into sufficiently distinct prognostic groups that potentially could be used to guide selection of adjuvant treatment by avoiding therapy in very low risk groups and identifying patients who would benefit from more intensive treatment regimes.

• Kleppe, Andreas; Albregtsen, Fritz; Trovik, Jone; Kristensen, Gunnar S Balle & Danielsen, Håvard Emil (2020). Prognostic value of the diversity of nuclear chromatin compartments in gynaecological carcinomas. Cancers. 12:3838(12), p. 1–16. doi: 10.3390/cancers12123838. Full text in Research Archive Show summary

  Statistical texture analysis of cancer cell nuclei stained for DNA has recently been used to develop a pan-cancer prognostic marker of chromatin heterogeneity. In this study, we instead analysed chromatin organisation by automatically quantifying the diversity of chromatin compartments in cancer cell nuclei. The aim was to investigate the prognostic value of such an assessment in relation to chromatin heterogeneity and as a potential supplement to pathological risk classifications in gynaecological carcinomas. The diversity was quantified by calculating the entropy of both chromatin compartment sizes and optical densities within compartments. We analysed a median of 281 nuclei (interquartile range (IQR), 273 to 289) from 246 ovarian carcinoma patients and a median of 997 nuclei (IQR, 502 to 1452) from 791 endometrial carcinoma patients. The prognostic value of the entropies and chromatin heterogeneity was moderately strongly correlated (r ranged from 0.68 to 0.73), but the novel marker was observed to provide additional prognostic information. In multivariable analysis with clinical and pathological markers, the hazard ratio associated with the novel marker was 2.1 (95% CI, 1.3 to 3.5) in ovarian carcinoma and 2.4 (95% CI, 1.5 to 3.9) in endometrial carcinoma. Integration with pathological risk classifications gave three risk groups with distinctly different prognoses. This suggests that the novel marker of diversity of chromatin compartments might possibly contribute to the selection of high-risk stage I ovarian carcinoma patients for adjuvant chemotherapy and low-risk endometrial carcinoma patients for less extensive surgery.

• Ersvær, Elin; Hveem, Tarjei Sveinsgjerd; Vlatkovic, Ljiljana; Brennhovd, Bjørn; Kleppe, Andreas & Tobin, Kari Anne Risan [Show all 11 contributors for this article] (2019). Prognostic value of DNA ploidy and automated assessment of stroma fraction in prostate cancer. International Journal of Cancer. ISSN 0020-7136. doi: 10.1002/ijc.32832. Full text in Research Archive
• Ersvær, Elin; Kildal, Wanja; Vlatkovic, Ljiljana; Cyll, Karolina Urszula; Pradhan, Manohar & Kleppe, Andreas [Show all 12 contributors for this article] (2019). Prognostic value of mitotic checkpoint protein BUB3, cyclin B1, and pituitary tumor-transforming 1 expression in prostate cancer. Modern Pathology. ISSN 0893-3952. p. 1–11. doi: 10.1038/s41379-019-0418-2. Full text in Research Archive
• Danielsen, Håvard Emil; Hveem, Tarjei Sveinsgjerd; Domingo, Enric; Pradhan, Manohar; Kleppe, Andreas & Syvertsen, Rolf Anders [Show all 18 contributors for this article] (2018). Prognostic markers for colorectal cancer: Estimating ploidy and stroma. Annals of Oncology. ISSN 0923-7534. 29(3), p. 616–623. doi: 10.1093/annonc/mdx794. Full text in Research Archive
• Nielsen, Birgitte; Kleppe, Andreas; Hveem, Tarjei Sveinsgjerd; Pradhan, Manohar; Syvertsen, Rolf Anders & Nesheim, John Arne [Show all 11 contributors for this article] (2018). Association Between Proportion of Nuclei With High Chromatin Entropy and Prognosis in Gynecological Cancers. Journal of the National Cancer Institute. ISSN 0027-8874. 110(12), p. 1400–1408. doi: 10.1093/jnci/djy063. Full text in Research Archive Show summary

  Background: Nuclear texture analysis measuring differences in chromatin structure has provided prognostic biomarkers in several cancers. There is a need for improved cell-by-cell chromatin analysis to detect nuclei with highly disorganized chromatin. The purpose of this study was to develop a method for detecting nuclei with high chromatin entropy and to evaluate the association between the presence of such deviating nuclei and prognosis. Methods: A new texture-based biomarker that characterizes each cancer based on the proportion of high–chromatin entropy nuclei (<25% vs ≥25%) was developed on a discovery set of 175 uterine sarcomas. The prognostic impact of this biomarker was evaluated on a validation set of 179 uterine sarcomas, as well as on independent validation sets of 246 early-stage ovarian carcinomas and 791 endometrial carcinomas. More than 1 million images of nuclei stained for DNA were included in the study. All statistical tests were two-sided. Results: An increased proportion of high–chromatin entropy nuclei was associated with poor clinical outcome. The biomarker predicted five-year overall survival for uterine sarcoma patients with a hazard ratio (HR) of 2.02 (95% confidence interval [CI] = 1.43 to 2.84), time to recurrence for ovarian cancer patients (HR = 2.91, 95% CI = 1.74 to 4.88), and cancer-specific survival for endometrial cancer patients (HR = 3.74, 95% CI = 2.24 to 6.24). Chromatin entropy was an independent prognostic marker in multivariable analyses with clinicopathological parameters (HR = 1.81, 95% CI = 1.21 to 2.70, for sarcoma; HR = 1.71, 95% CI = 1.01 to 2.90, for ovarian cancer; and HR = 2.03, 95% CI = 1.19 to 3.45, for endometrial cancer). Conclusions: A novel method detected high–chromatin entropy nuclei, and an increased proportion of such nuclei was associated with poor prognosis. Chromatin entropy supplemented existing prognostic markers in multivariable analyses of three gynecological cancer cohorts.

• Kleppe, Andreas; Albregtsen, Fritz; Vlatkovic, Ljiljana; Pradhan, Manohar; Nielsen, Birgitte & Hveem, Tarjei Sveinsgjerd [Show all 16 contributors for this article] (2018). Chromatin organisation and cancer prognosis: a pan-cancer study. The Lancet Oncology. ISSN 1470-2045. 19(3), p. 356–369. doi: 10.1016/S1470-2045(17)30899-9. Full text in Research Archive Show summary

  Background: Chromatin organisation affects gene expression and regional mutation frequencies and contributes to carcinogenesis. Aberrant organisation of DNA has been correlated with cancer prognosis in analyses of the chromatin component of tumour cell nuclei using image texture analysis. As yet, the methodology has not been sufficiently validated to permit its clinical application. We aimed to define and validate a novel prognostic biomarker for the automatic detection of heterogeneous chromatin organisation. Methods Machine learning algorithms analysed the chromatin organisation in 461 000 images of tumour cell nuclei stained for DNA from 390 patients (discovery cohort) treated for stage I or II colorectal cancer at the Aker University Hospital (Oslo, Norway). The resulting marker of chromatin heterogeneity, termed Nucleotyping, was subsequently independently validated in six patient cohorts: 442 patients with stage I or II colorectal cancer in the Gloucester Colorectal Cancer Study (UK); 391 patients with stage II colorectal cancer in the QUASAR 2 trial; 246 patients with stage I ovarian carcinoma; 354 patients with uterine sarcoma; 307 patients with prostate carcinoma; and 791 patients with endometrial carcinoma. The primary outcome was cancer-specific survival. Findings: In all patient cohorts, patients with chromatin heterogeneous tumours had worse cancer-specific survival than patients with chromatin homogeneous tumours (univariable analysis hazard ratio [HR] 1·7, 95% CI 1·2–2·5, in the discovery cohort; 1·8, 1·0–3·0, in the Gloucester validation cohort; 2·2, 1·1–4·5, in the QUASAR 2 validation cohort; 3·1, 1·9–5·0, in the ovarian carcinoma cohort; 2·5, 1·8–3·4, in the uterine sarcoma cohort; 2·3, 1·2–4·6, in the prostate carcinoma cohort; and 4·3, 2·8–6·8, in the endometrial carcinoma cohort). After adjusting for established prognostic patient characteristics in multivariable analyses, Nucleotyping was prognostic in all cohorts except for the prostate carcinoma cohort (HR 1·7, 95% CI 1·1–2·5, in the discovery cohort; 1·9, 1·1–3·2, in the Gloucester validation cohort; 2·6, 1·2–5·6, in the QUASAR 2 cohort; 1·8, 1·1–3·0, for ovarian carcinoma; 1·6, 1·0–2·4, for uterine sarcoma; 1·43, 0·68–2·99, for prostate carcinoma; and 1·9, 1·1–3·1, for endometrial carcinoma). Chromatin heterogeneity was a significant predictor of cancer-specific survival in microsatellite unstable (HR 2·9, 95% CI 1·0–8·4) and microsatellite stable (1·8, 1·2–2·7) stage II colorectal cancer, but microsatellite instability was not a significant predictor of outcome in chromatin homogeneous (1·3, 0·7–2·4) or chromatin heterogeneous (0·8, 0·3–2·0) stage II colorectal cancer. Interpretation: The consistent prognostic prediction of Nucleotyping in different biological and technical circumstances suggests that the marker of chromatin heterogeneity can be reliably assessed in routine clinical practice and could be used to objectively assist decision making in a range of clinical settings. An immediate application would be to identify high-risk patients with stage II colorectal cancer who might have greater absolute benefit from adjuvant chemotherapy. Clinical trials are warranted to evaluate the survival benefit and cost-effectiveness of using Nucleotyping to guide treatment decisions in multiple clinical settings.

• Cyll, Karolina Urszula; Ersvær, Elin; Vlatkovic, Ljiljana; Pradhan, Manohar; Kildal, Wanja & Kjær, Marte Avranden [Show all 15 contributors for this article] (2017). Tumour heterogeneity poses a significant challenge to cancer biomarker research. British Journal of Cancer. ISSN 0007-0920. 117(3), p. 367–375. doi: 10.1038/bjc.2017.171. Full text in Research Archive
• Hveem, Tarjei Sveinsgjerd; Kleppe, Andreas; Vlatkovic, Ljiljana; Ersvær, Elin; Wæhre, Haakon & Nielsen, Birgitte [Show all 13 contributors for this article] (2016). Chromatin changes predict recurrence after radical prostatectomy. British Journal of Cancer. ISSN 0007-0920. 114(11), p. 1243–1250. doi: 10.1038/bjc.2016.96. Full text in Research Archive

View all works in NVA

• Bakker, M-C.E.; Franken, I.A.; Laclé, M.M.; Cyll, Karolina; Stathonikos, N. & Raedt, Sepp de [Show all 17 contributors for this article] (2025). 726O Prognostic value of the combined analysis of pathologists and artificial intelligence (CAPAI) in high-risk stage II-III colon cancer treated without chemotherapy: Interim report from a nationwide validation. Annals of Oncology. ISSN 0923-7534. 36(Supplement 2), p. S476–S476. doi: 10.1016/j.annonc.2025.08.1301. Full text in Research Archive Show summary

  Background The recommended treatment of high-risk stage II (pT4) and stage III (pN+) colon cancer (CC) is surgical resection followed by adjuvant chemotherapy (ACT). However, half of patients are cured by surgery alone and may be spared ACT if identified up-front. The Combined Analysis of Pathologists and Artificial Intelligence (CAPAI) was developed and validated to predict cancer-specific survival (CSS) in patients predominantly treated with ACT. The restrictive use of ACT in the Netherlands enables validation of the prognostic value of CAPAI when withholding ACT. Methods We identified all patients from the Netherlands Cancer Registry (NCR) with high-risk stage II and III CC with R0 resection between 2015-2019, without (neo)adjuvant treatment, under 70 years of age and with a good performance status (WHO0-1 or ASA1-2). Diagnostic H&E slides were retrieved from the nationwide database (Palga), centrally scanned, and analyzed by the DoMore-v1 deep learning biomarker. The score was combined with pT/pN stage and lymph node count to classify patients into CAPAI risk groups: low-, intermediate-, or high-risk. The primary endpoint was 3-year CSS, defined as death with known recurrence. Results This interim analysis includes 453 patients from 65 hospitals, with an overall 3-year CSS of 85.9% (95% CI 82.6-89.2%). CAPAI stratified patients in significantly distinct prognostic groups (log rank p<0.0001) with 3-year CSS of 93.7% (90.5-97.1%; n=215 [47%]) for low-, 87.5% (82.3-93.1; n=156 [34%]) for intermediate- and 60.4% (50.4-72.5; n=82 [18%]) for high-risk. In univariable Cox regression, CAPAI provided a hazard ratio for high- versus low-risk of 7.9 (4.1-15.2; p < 0.001) and for intermediate- versus low-risk of 2.1 (1.0-4.3; p=0.039). Conclusions CAPAI successfully stratified patients into distinct prognostic groups, identifying nearly half of patients as low-risk with a 3-year CSS of 93.7% without ACT. A comparison with a balanced cohort of patients who received ACT is ongoing. Prospective use of this readily available and affordable stratification tool could help avoid ACT in low-risk patients and intensify treatment in high-risk patients. Legal entity responsible for the study The authors. Funding CAPAI is developed thanks to funding by Learning from Deep learning, The Norwegian Research Council (#334862) and Improving diagnosis by utilizing Big Data and software-driven automation of pathology, The Norwegian Research Council (#259204). This public private partnership was sponsored by Health∼Holland (LSHM#19027).

• Lujan, G.; Arole, V.; Raedt, Sepp de; Meyers, K.; Pradhan, Manohar & Maupin, J. [Show all 17 contributors for this article] (2025). 52P External validation of Histotype Px® Colorectal for risk stratification in a US patient cohort. doi: https:/www.sciencedirect.com/science/article/pii/S2949820125001407?via%3Dihub. Full text in Research Archive Show summary

  Background Many patients with stage II/III colorectal cancer (CRC) are routinely offered adjuvant chemotherapy (ACT) according to guidelines. However, with significant side-effects and few patients benefitting, better risk stratification is needed. Histotype Px Colorectal is an AI-based digital biomarker that analyses routine H&E whole-slide images (WSI) and combined with pT, pN, and number of lymph nodes sampled, stratifies patients into distinct risk groups. We externally validated its prognostic performance. Methods Patients diagnosed with stage II/III colon adenocarcinoma at Ohio State University from 2011-2024 were included in this retrospective study. Blinded to clinical outcomes, anonymized routine Philips iSyntax WSIs of H&E sections were analysed by Histotype Px. Logrank test and Cox proportional hazards regression were subsequently used to analyse cancer-specific survival (CSS) and time to recurrence (TTR) to determine the prognostic performance. Results A total of 503 patients were included with a median age of 63 years (22-97), 54% male, 45% stage II, 61% right-sided tumors, 23% MSI, 51% received ACT, and 44 months median follow-up. Histotype Px was a significant predictor of CSS with hazard ratio (HR) 8.7 (95% CI 4.4-17.3, p<0.001) for high vs. low risk patients and HR 2.9 (95% CI 1.4-6.0, p=0.005) for intermediate vs. low risk patients. Histotype Px was a significant predictor of TTR with HR 4.5 (2.9-6.9, p<0.001) for high vs. low risk patients and HR 1.8 (1.2-2.9, p=0.008) for intermediate vs. low risk patients. Only 11 of 228 stage II patients had a cancer-related death, corresponding with 96% and 84% 5-year CSS for low and intermediate risk, respectively. For stage III patients, 41% were classified as low, 29% as intermediate, and 30% as high risk. 40 of the 279 stage III patients had a cancer-related death, with 5-year CSS of 92%, 83%, and 63% for low, intermediate, and high risk, respectively. Conclusions Histotype Px Colorectal is an AI-based digital biomarker that stratifies stage II/III patients into distinct risk groups. It identifies patients with exceedingly low risk of cancer-specific death who may potentially forgo ACT, and conversely, patients who may benefit from treatment escalation, thereby enabling more personalized ACT for CRC. Legal entity responsible for the study Ohio State University Comprehensive Cancer Center. Funding Ohio State University Comprehensive Cancer Center.

• Pradhan, Manohar; Kildal, Wanja; Kleppe, Andreas; Tobin, Kari Anne Risan; Nikolaysen, Filip & Mascialino, Barbara [Show all 12 contributors for this article] (2025). Prognostic Value Of Different Stages Of Maturation Of Tertiary Lymphoid Structures In Molecular Subgroups Of Endometrial Carcinoma. International Journal of Gynecological Cancer. ISSN 1048-891X. 35(2), p. 100104–100104. doi: 10.1016/j.ijgc.2024.100104. Full text in Research Archive Show summary

  Introduction/Background Tertiary lymphoid structures (TLSs) are organized clusters of lymphoid cells at various maturation stages with potential prognostic and predictive impact for response to immunotherapy in several cancer types. We assessed the prognostic significance of TLSs in patients with molecularly classified endometrial carcinoma (EC). Methodology This retrospective cohort included 1221 patients with molecularly classified EC treated at Oslo University Hospital 2006-2017. Formalin-fixed, paraffin-embedded tissue sections from hysterectomy specimens were immunohistochemically stained for CD21, Ki-67, CD3 and CD20. TLSs were categorized as mature (CD21+/Ki-67+), immature (CD21+/Ki-67–), lymphoid aggregates (CD21–, CD3+, and CD20+), or negative (CD21–/no CD3/CD20+ clusters). Univariable and multivariable analyses were performed to assess the association between TLS stages and time to recurrence (TTR). Results We identified 517 patients with lymphoid aggregates, 224 with immature TLSs, 138 with mature TLSs, and 342 TLS-negative cases. TLSs were more frequent in POLE-mutated (98%, 94/96) and MMRd (MMR deficient) tumours (87%, 321/368) compared to p53 abnormal (63%, 175/275) and no specific molecular profile (60%, 278/462) tumours (p<0.001). TLS presence of any maturity was significantly associated with longer TTR in univariable (p<0.001) and multivariable analysis (p<0.001). Compared to TLS-negative cases, the hazard ratio was 0.69 (95% confidence interval (95%CI): 0.54-0.89) for lymphoid aggregates, 0.45 (95%CI: 0.31-0.64) for immature TLSs, and 0.27 (95%CI: 0.16-0.44) for mature TLSs in multivariable analyses. In stratified analysis by molecular subgroup, presence of all maturity stages of TLSs remained significantly associated with longer TTR in patients in MMRd and NSMP tumours in multivariable analysis. TLSs is a favourable prognostic marker for EC patients, especially in the MMRd and NSMP subgroups. Conclusion TLSs were predominantly found in patients with POLE-mutated and MMRd tumours. Prognosis improved progressively with TLS maturation stage, with longest TTR in patients with mature TLSs. Evaluating TLSs may provide additional prognostic information across molecular subgroups. Disclosures This supported collaborative study was funded by GSK (OSF.IO/JHEVG” title="https://doi.org/10.17605/OSF.IO/JHEVG">https://doi.org/10.17605/OSF.IO/JHEVG). BM, DS, and HE are full-time employees of GSK and hold stock and/or stock options. HE holds stock/stock options i Abbvie. KL reports the following conflicts of interest outside the submitted work: Participation on data safety monitoring or advisory boards of Eisai, MSD, Nykode, AstraZeneca, GSK and Karyopharm (honoraria paid to institution); speaker fee from MSD; GSK and Astra Zeneca and research funding paid to institution from GSK. All remaining authors have declared no conflicts of interest.

• Cyll, Karolina; Skrede, Ole-Johan & Kleppe, Andreas (2025). Can radiology be first to use prognostic deep learning models for oncological treatment? Annals of Oncology. ISSN 0923-7534. 36(10), p. 1122–1122. doi: 10.1016/j.annonc.2025.07.013. Full text in Research Archive
• Pradhan, Manohar; Kildal, Wanja; Vlatkovic, Ljiljana; Tobin, Kari Anne Risan; Lindemann, Kristina Yvonne Kathe & Kristensen, Gunnar S Balle [Show all 8 contributors for this article] (2025). Response to correspondence on prognostic value of tertiary lymphoid structures in molecular subgroups of endometrial carcinoma by Pradhan et al. International Journal of Gynecological Cancer. ISSN 1048-891X. 35(7), p. 101970–101970. doi: 10.1016/j.ijgc.2025.101970. Full text in Research Archive
• Henriksen, Audun Ljone; Julbø, Frida Marie Ihle; Pradhan, Manohar; Lindstrøm, Erika Kristina; Schee, Lisa van der & Isaksen, Maria [Show all 15 contributors for this article] (2024). E-PS-17-037 Improving histopathological screening of colorectal polyps using deep learning. Virchows Archiv. ISSN 0945-6317. 485(Suppl 1), p. S389–S389. Full text in Research Archive Show summary

  Background & objectives: Few polyps harvested during colonoscopy warrant surgery or additional surveillance. Automatic identification of polyps that are low risk would reduce the workload on pathologists. We aimed to develop such a marker by applying deep learning to analyse histopathological section images. Methods: The DoMore-v1 network was modified to perform attentionbased pooling and trained to classify histopathological section images as 1)normal tissue, 2)tubular adenoma low-grade dysplasia, 3)sessile serrated lesion or hyperplastic polyp, 4)high-grade dysplasia or adenocarcinoma, or 5)other. It was developed using 11601 histopathological section images from 2361 patients from United Kingdom and externally tested on 469 images from 295 patients from the The Netherlands. Results: In the external test, 58 (54.2%) of 107 histopathological section images labelled as tubular adenoma low-grade dysplasia were correctly classified and 1 (0.3%) of 301 images labelled as high-grade dysplasia or adenocarcinoma was misclassified as normal tissue or tubular adenoma low-grade dysplasia, further collectively called lowrisk polyps. Three pathologists independently and blindly reviewed this misclassified image, and two of the pathologists classified it as tubular adenoma low-grade dysplasia. For the remaining 61 images in the external test dataset, the biomarker correctly classified 59 (96.7%) of them as not being low-risk polyps. Conclusion: External testing of a new deep learning marker indicated that more than half of the low-risk polyps can be automatically identified without incorrectly identifying polyps with high-grade dysplasia or adenocarcinoma as being of low-risk type. Allowing pathologists to not assess the automatically identified low-risk polyps will make bowel cancer screening more cost-effective and feasible by substantially reducing the workload for pathologists classifying colorectal polyps. Funding: This work was supported by the Norwegian Cancer Society (grant number 215919) and The Research Council of Norway through its IKTPLUSS Lighthouse program (grant number 259204).

• Pradhan, Manohar; Kildal, Wanja; Vlatkovic, Ljiljana; Lindemann, Kristina Yvonne Kathe; Tobin, Kari Anne Risan & Kristensen, Gunnar S Balle [Show all 8 contributors for this article] (2024). OFP-02-005 Prognostic value of tertiary lymphoid structures in endometrial carcinoma. Virchows Archiv. ISSN 0945-6317. 485(S1), p. S6–S7. doi: 10.1007/s00428-024-03880-y. Full text in Research Archive Show summary

  Background & objectives: Tertiary lymphoid structures (TLS), organized aggregates of lymphoid cells, are reported as an immunological prognostic marker. We have evaluated the prognostic importance of TLS in patients with endometrial carcinoma using L1 cell adhesion molecule (L1CAM) as a surrogate marker. Methods: The cohort consists of 1228 endometrial carcinoma patients (880 endometrioid and 348 non-endometrioid) treated at Oslo University Hospital between 2006 and 2017. Formalin-fixed and paraffin- embedded tissue samples from the hysterectomy specimens were sectioned and one section was immunohistochemically stained with antibody against L1CAM to evaluate the presence and number of TLS. Sections with ≥1 TLS were considered as TLS positive. Results: Out of 1197 evaluable sections, 911 (76%) sections were TLS negative, while 286 (24%) were TLS positive. In TLS positive sections, the median number was 3 (IQR 1-6). TLS positive patients had a significantly lower risk of recurrence (HR 0.60; 95% CI 0.45-0.8, p<0.001) compared to TLS negative patients in univariable analyses. In molecular subgroups, TLS were more frequently present in tumours with POLE mutation (59%) and MMR deficiency (32%) than in tumours with p53 abnormality (16%) and no specific molecular profile (15%). In patients with MMR deficiency and p53 abnormal tumours, presence of TLS was prognostic with HR of 0.6 (CI 0.37-0.98, p=0.042) and 0.6 (CI 0.68-1.99, p=0.056), respectively. Conclusion: Evaluation of TLS using L1CAM might provide additional prognostic information for endometrial carcinoma patients with MMR deficiency. A similar indication was observed for patients with p53 abnormality, but further studies are needed to confirm the prognostic value in this subgroup. TLS were predominantly found in POLE mutated and MMR deficient tumours. These findings corroborate with a previous study investigating the correlation and prognostic value of TLS in molecular subgroups using a much smaller patient cohort.

• Lindemann, Kristina Yvonne Kathe; Kildal, Wanja; Kleppe, Andreas; Tobin, Kari Anne Risan; Pradhan, Manohar & Isaksen, Maria [Show all 10 contributors for this article] (2024). 259 Impact of molecular profile on prognosis and relapse pattern in low and intermediate risk endometrial cancer. International Journal of Gynecological Cancer. ISSN 1048-891X. 34(Supplement 1), p. A167–A168. doi: 10.1136/ijgc-2024-esgo.317. Full text in Research Archive Show summary

  Introduction/Background The role of molecular classification in patients with low/intermediate risk endometrial cancer (EC) is uncertain. Higher precision in diagnostics will inform the unsettled debate on optimal adjuvant treatment. Methodology We retrospectively included patients referred to The Norwegian Radium Hospital, Oslo, Norway (2006–2017). Patients with low/intermediate risk EC were molecularly classified as pathogenic polymerase epsilon (POLE)-mutated, mismatch repair deficient (MMRd), p53 abnormal, or no specific molecular profile (NSMP). The main outcome was time to recurrence (TTR) and cancer-specific survival (CSS) within the molecular subgroups. We also studied patterns of recurrence in molecular groups. Results Of 626 patients, 610 could be molecularly classified and only 24 received adjuvant treatment. Molecular subgroup distribution was: 57 patients (9%) with POLE-mutated tumors, 202 (33%) with MMRd tumors, 34 (6%) with p53 abnormal tumors, and 317 (52%) with NSMP tumors. After a median follow-up time of 8.9 years (95% confidence interval (CI): 6.2–12.6 years), there was a statistically significant difference in TTR (p<0.001) and CSS (p<0.001) by molecular groups. Patients with p53 abnormal tumors had poor prognosis with a 5-year cumulative incidence for recurrence of 32.4% (95% CI: 17.6–48.0) and cancer-specific death (CSD) of 29.4% (95% CI: 15.4–44.9). The majority of these recurrences were para-aortic/distant metastases. Patients with POLE mutations had excellent prognosis with a 5-year cumulative incidence for recurrence of 3.5% (95% CI: 0.7–10.7%) and CSD of 0%. In the NSMP group, L1CAM expression was associated with shorter CSS (HR 12.52, 95% CI: 3.53–44.42, p<0.0001) but not TTR (HR 2.7, 95% CI: 0.95–7.7, p=0.053). Conclusion Large differences in prognosis and localization of recurrence by molecular groups demand a higher precision in diagnostics, also in patients with low and intermediate risk EC. Tailored adjuvant treatment strategies need to consider systemic treatment for patients with p53 abnormal tumors and de-escalation of treatment in patients with POLE mutated tumors. Disclosures KL: Receipt of grants/research supports: GSK, research funding paid to Institution; Receipt of honoraria or consultation fees: Advisory board fees: Astra Zeneca, Nycode, GSK, Eisa GK: Stockholder: Novo Nordic

• Lindemann, Kristina Yvonne Kathe; Kildal, Wanja; Kleppe, Andreas; Tobin, Kari Anne Risan; Pradhan, Manohar & Mascialino, Barbara [Show all 13 contributors for this article] (2024). 258 Real-world data of patients with stage III/IV endometrial cancer treated with platinum-based chemotherapy after hysterectomy – outcome in molecular subgroups. International Journal of Gynecological Cancer. ISSN 1048-891X. 34 (supplement 1), p. A167–A167. doi: 10.1136/ijgc-2024-esgo.316. Full text in Research Archive Show summary

  Introduction/Background There is scarce real-world evidence on patients with advanced endometrial cancer (EC) treated with platinum-based chemotherapy in the literature. This study described the oncological outcome in groups by molecular classification. Methodology This retrospective cohort study included patients referred to The Norwegian Radium Hospital, Oslo University Hospital (OUH), Norway who were diagnosed or operated for EC at OUH between January 2006 and December 2017. Patients with Stage III/IV EC treated with platinum-based chemotherapy after hysterectomy and with sufficient tumor tissue available at the biobank were molecularly classified as pathogenic polymerase epsilon (POLE)-mutated, mismatch repair deficient (MMRd), p53 abnormal, or no specific molecular profile (NSMP). Results Of 264 patients, 263 were molecularly classified, 10 (4%) as POLE-mutated, 67 (26%) as MMRd, 117 (44%) as p53 abnormal and 69 (26%) as NSMP. One patient could not be molecularly characterized. Median age was 68 (61–74) years with a median follow up time of 5.3 (2.2–10.3) years. Patients with MMRd tumors had significantly longer time to recurrence (HR 0.43, 95% CI: 0.27–0.67, p=0.0001) and longer cancer-specific survival (HR 0.36, 95%CI: 0.22–0.58, p<0.0001) than patients with mismatch repair proficient (MMRp) tumors. Compared to patients with NSMP tumors, patients with p53 abnormal tumors had shorter cancer-specific survival (HR 1.78, 95% CI: 1.19–2.65) while patients with MMRd tumors had longer cancer-specific survival (HR 0.50, 95% CI: 0.29–0.87). Patients with POLE tumors had the longest cancer-specific survival (HR 0.32, 95% 0.08–1.32). Conclusion Results show that molecular classification is prognostic in patients with advanced-stage endometrial cancer. Based on the recently reported clinical trials of a checkpoint inhibitor added to platinum-based chemotherapy in advanced and recurrent EC, patients with MMRd, and maybe also p53 abnormal tumors, have a better clinical outcome when treated with this combination treatment Disclosures This supported collaborative study was funded by GSK (SCS: 219026) KL: Receipt of grants/research supports: GSK, research funding paid to Institution; Receipt of honoraria or consultation fees: Advisory board fees: Astra Zeneca, Nycode, GSK, Eisa BM, DS: Stockholder: GSK, Other: Employee GSK HE: Other: Employee GSK GK: Stockholder: Novo Nordic

• Kleppe, Andreas; Skrede, Ole-Johan; Liestøl, Knut; Kerr, David J. & Danielsen, Håvard Emil Greger (2024). Guidelines for study protocols describing predefined validations of prediction models in medical deep learning and beyond. Nature Machine Intelligence. 6(1), p. 2–3. doi: 10.1038/s42256-023-00774-7. Full text in Research Archive
• Kleppe, Andreas (2022). Area under the curve may hide poor generalisation to external datasets. ESMO Open. 7(2). doi: 10.1016/j.esmoop.2022.100429. Full text in Research Archive
• Kleppe, Andreas (2021). Designing deep learning studies in medical diagnostics and beyond . Full text in Research Archive
• Liestøl, Knut; Kleppe, Andreas; Skrede, Ole-Johan; Raedt, Sepp De; Kerr, David J. & Danielsen, Håvard Emil Greger (2021). Deep learning and cancer biomarkers: recognising lead-time bias – Authors' reply. The Lancet. ISSN 0140-6736. 397(10270), p. 194–195. doi: 10.1016/S0140-6736(21)00035-0. Full text in Research Archive
• Kleppe, Andreas & Danielsen, Håvard Emil (2018). Clinical utility of chromatin analysis. OncoTarget. 9(65), p. 32406–32407. doi: 10.18632/oncotarget.25991. Full text in Research Archive

View all works in NVA