Anca Dragomir
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Anca Dragomir has a total of 184 co-authors. Here are the top co-authors by number of shared publications:
Godfried M. Roomans
School of Health and Medical Sciences, Örebro University, Örebro, Sweden
10 shared publications- Per Pfeiffer
Department of Oncology, Odense University Hospital, Odense, Denmark; Department of Clinical Research, University of Southern Denmark, Odense, Denmark
7 shared publications Bengt Glimelius
Department of Immunology, Genetics and Pathology, Uppsala University, Uppsala, Sweden
6 shared publicationsHalfdan Sorbye
Department of Clinical Science, University of Bergen, Bergen, Norway
6 shared publications- Artur Mezheyeuski
Vall d’Hebron Institute of Oncology, 08035 Barcelona, Spain
5 shared publications - Harriet Elisabeth Nilsson
Department of Biosciences and Nutrition, Karolinska Institutet, and School of Technology and Health, KTH Royal Institute of Technology, Novum, SE-141 57 Huddinge, Sweden
5 shared publications
Latest Publications
- Open Access
- 21 November 2025
- Published by Elsevier in Medical Image Analysis
- Vol. 108, 103884
A systematic analysis of the impact of data variation on AI-based histopathological grading of prostate cancer
- 21 November 2025
- Published by Elsevier in Medical Image Analysis
- Vol. 108, 103884
Open AccessA systematic analysis of the impact of data variation on AI-based histopathological grading of prostate cancer
- Open Access
Robust, credible, and interpretable AI-based histopathological prostate cancer grading
Background Prostate cancer (PCa) is among the most common cancers in men and its diagnosis requires the histopathological evaluation of biopsies by human experts. While several recent artificial intelligence-based (AI) approaches have reached human expert-level PCa grading, they often display significantly reduced performance on external datasets. This reduced performance can be caused by variations in sample preparation, for instance the staining protocol, section thickness, or scanner used. Another limiting factor of contemporary AI-based PCa grading is the prediction of ISUP grades, which leads to the perpetuation of human annotation errors. Methods We developed the prostate cancer aggressiveness index (PCAI), an AI-based PCa detection and grading framework that is trained on objective patient outcome, rather than subjective ISUP grades. We designed PCAI as a clinical application, containing algorithmic modules that offer robustness to data variation, medical interpretability, and a measure of prediction confidence. To train and evaluate PCAI, we generated a multicentric, retrospective, observational trial consisting of six cohorts with 25,591 patients, 83,864 images, and 5 years of median follow-up from 5 different centers and 3 countries. This includes a high-variance dataset of 8,157 patients and 28,236 images with variations in sample thickness, staining protocol, and scanner, allowing for the systematic evaluation and optimization of model robustness to data variation. The performance of PCAI was assessed on three external test cohorts from two countries, comprising 2,255 patients and 9,437 images. Findings Using our high-variance datasets, we show how differences in sample processing, particularly slide thickness and staining time, significantly reduce the performance of AI-based PCa grading by up to 6.2 percentage points in the concordance index (C-index). We show how a select set of algorithmic improvements, including domain adversarial training, conferred robustness to data variation, interpretability, and a measure of credibility to PCAI. These changes lead to significant prediction improvement across two biopsy cohorts and one TMA cohort, systematically exceeding expert ISUP grading in C-index and AUROC by up to 22 percentage points. Interpretation Data variation poses serious risks for AI-based histopathological PCa grading, even when models are trained on large datasets. Algorithmic improvements for model robustness, interpretability, credibility, and training on high-variance data as well as outcome-based severity prediction gives rise to robust models with above ISUP-level PCa grading performance.
Open AccessRobust, credible, and interpretable AI-based histopathological prostate cancer grading
Background Prostate cancer (PCa) is among the most common cancers in men and its diagnosis requires the histopathological evaluation of biopsies by human experts. While several recent artificial intelligence-based (AI) approaches have reached human expert-level PCa grading, they often display significantly reduced performance on external datasets. This reduced performance can be caused by variations in sample preparation, for instance the staining protocol, section thickness, or scanner used. Another limiting factor of contemporary AI-based PCa grading is the prediction of ISUP grades, which leads to the perpetuation of human annotation errors. Methods We developed the prostate cancer aggressiveness index (PCAI), an AI-based PCa detection and grading framework that is trained on objective patient outcome, rather than subjective ISUP grades. We designed PCAI as a clinical application, containing algorithmic modules that offer robustness to data variation, medical interpretability, and a measure of prediction confidence. To train and evaluate PCAI, we generated a multicentric, retrospective, observational trial consisting of six cohorts with 25,591 patients, 83,864 images, and 5 years of median follow-up from 5 different centers and 3 countries. This includes a high-variance dataset of 8,157 patients and 28,236 images with variations in sample thickness, staining protocol, and scanner, allowing for the systematic evaluation and optimization of model robustness to data variation. The performance of PCAI was assessed on three external test cohorts from two countries, comprising 2,255 patients and 9,437 images. Findings Using our high-variance datasets, we show how differences in sample processing, particularly slide thickness and staining time, significantly reduce the performance of AI-based PCa grading by up to 6.2 percentage points in the concordance index (C-index). We show how a select set of algorithmic improvements, including domain adversarial training, conferred robustness to data variation, interpretability, and a measure of credibility to PCAI. These changes lead to significant prediction improvement across two biopsy cohorts and one TMA cohort, systematically exceeding expert ISUP grading in C-index and AUROC by up to 22 percentage points. Interpretation Data variation poses serious risks for AI-based histopathological PCa grading, even when models are trained on large datasets. Algorithmic improvements for model robustness, interpretability, credibility, and training on high-variance data as well as outcome-based severity prediction gives rise to robust models with above ISUP-level PCa grading performance.
- Open Access
- 14 March 2022
- Published by Taylor & Francis in OncoImmunology
- Vol. 11 (1) , 2049487
Neutrophil extracellular traps promote cancer-associated inflammation and myocardial stress
Cancer is associated with systemic pathologies that contribute to mortality, such as thrombosis and distant organ failure. The aim of this study was to investigate the potential role of neutrophil extracellular traps (NETs) in myocardial inflammation and tissue damage in treatment-naïve individuals with cancer. Mice with mammary carcinoma (MMTV-PyMT) had increased plasma levels of NETs measured as H3Cit-DNA complexes, paralleled with elevated coagulation, compared to healthy littermates. MMTV-PyMT mice displayed upregulation of pro-inflammatory markers in the heart, myocardial hypertrophy and elevated cardiac disease biomarkers in the blood, but not echocardiographic heart failure. Moreover, increased endothelial proliferation was observed in hearts from tumor-bearing mice. Removal of NETs by DNase I treatment suppressed the myocardial inflammation, expression of cardiac disease biomarkers and endothelial proliferation. Compared to a healthy control group, treatment-naïve cancer patients with different malignant disorders had increased NET formation, which correlated to plasma levels of the inflammatory marker CRP and the cardiac disease biomarkers NT-proBNP and sTNFR1, in agreement with the mouse data. Altogether, our data indicate that NETs contribute to inflammation and myocardial stress during malignancy. These findings suggest NETs as potential therapeutic targets to prevent cardiac inflammation and dysfunction in cancer patients.
- 14 March 2022
- Published by Taylor & Francis in OncoImmunology
- Vol. 11 (1) , 2049487
Open AccessNeutrophil extracellular traps promote cancer-associated inflammation and myocardial stress
Cancer is associated with systemic pathologies that contribute to mortality, such as thrombosis and distant organ failure. The aim of this study was to investigate the potential role of neutrophil extracellular traps (NETs) in myocardial inflammation and tissue damage in treatment-naïve individuals with cancer. Mice with mammary carcinoma (MMTV-PyMT) had increased plasma levels of NETs measured as H3Cit-DNA complexes, paralleled with elevated coagulation, compared to healthy littermates. MMTV-PyMT mice displayed upregulation of pro-inflammatory markers in the heart, myocardial hypertrophy and elevated cardiac disease biomarkers in the blood, but not echocardiographic heart failure. Moreover, increased endothelial proliferation was observed in hearts from tumor-bearing mice. Removal of NETs by DNase I treatment suppressed the myocardial inflammation, expression of cardiac disease biomarkers and endothelial proliferation. Compared to a healthy control group, treatment-naïve cancer patients with different malignant disorders had increased NET formation, which correlated to plasma levels of the inflammatory marker CRP and the cardiac disease biomarkers NT-proBNP and sTNFR1, in agreement with the mouse data. Altogether, our data indicate that NETs contribute to inflammation and myocardial stress during malignancy. These findings suggest NETs as potential therapeutic targets to prevent cardiac inflammation and dysfunction in cancer patients.
- Open Access
- 1 May 2020
- Published by Springer Nature in BMC Medical Genetics
- Vol. 21 (1) , 1-9
Rare variants in dynein heavy chain genes in two individuals with situs inversus and developmental dyslexia: a case report
Background: Developmental dyslexia (DD) is a neurodevelopmental learning disorder with high heritability. A number of candidate susceptibility genes have been identified, some of which are linked to the function of the cilium, an organelle regulating left-right asymmetry development in the embryo. Furthermore, it has been suggested that disrupted left-right asymmetry of the brain may play a role in neurodevelopmental disorders such as DD. However, it is unknown whether there is a common genetic cause to DD and laterality defects or ciliopathies.Case presentation: Here, we studied two individuals with co-occurringsitus inversus(SI) and DD using whole genome sequencing to identify genetic variants of importance for DD and SI. Individual 1 had primary ciliary dyskinesia (PCD), a rare, autosomal recessive disorder with oto-sino-pulmonary phenotype and SI. We identified two rare nonsynonymous variants in the dynein axonemal heavy chain 5 gene (DNAH5): a previously reported variant c.7502G > C; p.(R2501P), and a novel variant c.12043 T > G; p.(Y4015D). Both variants are predicted to be damaging. Ultrastructural analysis of the cilia revealed a lack of outer dynein arms and normal inner dynein arms. MRI of the brain revealed no significant abnormalities. Individual 2 had non-syndromic SI and DD. In individual 2, one rare variant (c.9110A > G;p.(H3037R)) in the dynein axonemal heavy chain 11 gene (DNAH11),coding for another component of the outer dynein arm, was identified.Conclusions: We identified the likely genetic cause of SI and PCD in one individual, and a possibly significant heterozygosity in the other, both involving dynein genes. Given the present evidence, it is unclear if the identified variants also predispose to DD and further studies into the association between laterality, ciliopathies and DD are needed.
- 1 May 2020
- Published by Springer Nature in BMC Medical Genetics
- Vol. 21 (1) , 1-9
Open AccessRare variants in dynein heavy chain genes in two individuals with situs inversus and developmental dyslexia: a case report
Background: Developmental dyslexia (DD) is a neurodevelopmental learning disorder with high heritability. A number of candidate susceptibility genes have been identified, some of which are linked to the function of the cilium, an organelle regulating left-right asymmetry development in the embryo. Furthermore, it has been suggested that disrupted left-right asymmetry of the brain may play a role in neurodevelopmental disorders such as DD. However, it is unknown whether there is a common genetic cause to DD and laterality defects or ciliopathies.Case presentation: Here, we studied two individuals with co-occurringsitus inversus(SI) and DD using whole genome sequencing to identify genetic variants of importance for DD and SI. Individual 1 had primary ciliary dyskinesia (PCD), a rare, autosomal recessive disorder with oto-sino-pulmonary phenotype and SI. We identified two rare nonsynonymous variants in the dynein axonemal heavy chain 5 gene (DNAH5): a previously reported variant c.7502G > C; p.(R2501P), and a novel variant c.12043 T > G; p.(Y4015D). Both variants are predicted to be damaging. Ultrastructural analysis of the cilia revealed a lack of outer dynein arms and normal inner dynein arms. MRI of the brain revealed no significant abnormalities. Individual 2 had non-syndromic SI and DD. In individual 2, one rare variant (c.9110A > G;p.(H3037R)) in the dynein axonemal heavy chain 11 gene (DNAH11),coding for another component of the outer dynein arm, was identified.Conclusions: We identified the likely genetic cause of SI and PCD in one individual, and a possibly significant heterozygosity in the other, both involving dynein genes. Given the present evidence, it is unclear if the identified variants also predispose to DD and further studies into the association between laterality, ciliopathies and DD are needed.
- Open Access
Rare variants in dynein heavy chain genes in two individuals with situs inversus and developmental dyslexia
Background: Developmental dyslexia (DD) is a neurodevelopmental learning disorder with high heritability. A number of candidate susceptibility genes have been identified, some of which are linked to the function of the cilium, an organelle regulating left-right asymmetry development in the embryo. Furthermore, it has been suggested that disrupted left-right asymmetry of the brain may play a role in neurodevelopmental disorders such as DD. Methods: Here, we studied two individuals with co-occurring situs inversus (SI) and DD using whole genome sequencing to identify single nucleotide variants or copy number variations of importance for DD and SI. Results: Individual 1 had primary ciliary dyskinesia (PCD), a rare, autosomal recessive disorder with oto-sino-pulmonary phenotype and SI. We identified two rare nonsynonymous variants in the dynein axonemal heavy chain 5 gene ( DNAH5) : c.7502G>C;p.(R2501P), a previously reported variant predicted to be damaging and c.12043T>G;p.(Y4015D), a novel variant predicted to be damaging. Ultrastructural analysis of the cilia revealed a lack of outer dynein arms and normal inner dynein arms. MRI of the brain revealed no significant abnormalities. Individual 2 had non-syndromic SI and DD. In individual 2, one rare variant (c.9110A>G;p.(H3037R)) in the dynein axonemal heavy chain 11 gene ( DNAH11) , coding for another component of the outer dynein arm, was identified. Conclusions: We identified the likely genetic cause of SI and PCD in one individual, and a possibly significant heterozygosity in the other, both involving dynein genes. Given the present evidence, it is unclear if the identified variants also predispose to DD, but further studies into the association are warranted.
Open AccessRare variants in dynein heavy chain genes in two individuals with situs inversus and developmental dyslexia
Background: Developmental dyslexia (DD) is a neurodevelopmental learning disorder with high heritability. A number of candidate susceptibility genes have been identified, some of which are linked to the function of the cilium, an organelle regulating left-right asymmetry development in the embryo. Furthermore, it has been suggested that disrupted left-right asymmetry of the brain may play a role in neurodevelopmental disorders such as DD. Methods: Here, we studied two individuals with co-occurring situs inversus (SI) and DD using whole genome sequencing to identify single nucleotide variants or copy number variations of importance for DD and SI. Results: Individual 1 had primary ciliary dyskinesia (PCD), a rare, autosomal recessive disorder with oto-sino-pulmonary phenotype and SI. We identified two rare nonsynonymous variants in the dynein axonemal heavy chain 5 gene ( DNAH5) : c.7502G>C;p.(R2501P), a previously reported variant predicted to be damaging and c.12043T>G;p.(Y4015D), a novel variant predicted to be damaging. Ultrastructural analysis of the cilia revealed a lack of outer dynein arms and normal inner dynein arms. MRI of the brain revealed no significant abnormalities. Individual 2 had non-syndromic SI and DD. In individual 2, one rare variant (c.9110A>G;p.(H3037R)) in the dynein axonemal heavy chain 11 gene ( DNAH11) , coding for another component of the outer dynein arm, was identified. Conclusions: We identified the likely genetic cause of SI and PCD in one individual, and a possibly significant heterozygosity in the other, both involving dynein genes. Given the present evidence, it is unclear if the identified variants also predispose to DD, but further studies into the association are warranted.
- Open Access
- 10 January 2020
- Published by Medical Journals Sweden AB in Acta Oncologica
- Vol. 59 (4) , 417-426
Molecular characterization of a large unselected cohort of metastatic colorectal cancers in relation to primary tumor location, rare metastatic sites and prognosis
Background: We have reported that BRAF V600E mutations and microsatellite instability-high (MSI-H) are more prevalent in a population-based cohort of metastatic colorectal cancer (mCRC) patients than has been reported from clinical trials or hospital-based patient groups. The aim was to explore if other mutations in mCRC differ in prevalence between these cohorts in relation to mismatch repair status and primary tumor location and if presence of bone or brain metastases is associated with any mutations. Material and methods: A population-based cohort of 798 mCRC patients from three regions in Scandinavia was used. Forty-four cancer related genes were investigated in a custom designed Ampliseq hotspot panel. Differences in survival were analyzed using the Kaplan–Meier estimator and the Cox regression analysis. Results: Determination of mutations was possible in 449/501 patients for 40/44 genes. Besides BRAF V600E, seen in 19% of the tumors, none of the other mutations appeared more prevalent than in trial cohorts. BRAF V600E and MSI-H, seen in 8%, were associated with poor prognosis as was right-sided primary tumor location (39%) when compared to left-sided and rectum together; however, in a multivariable regression, only the BRAF mutation retained its statistical significance. No other mutations were associated with poor prognosis. ERBB2 alterations were more common if bone metastases were present at diagnosis (17% vs. 4%, p = .011). No association was found for brain metastases. Fifty-two percent had an alteration that is treatable with an FDA-approved targeted therapy, chiefly by EGFR-inhibitor for RAS wild-type and a check-point inhibitor for MSI-H tumors. Conclusions: Right-sided tumor location, BRAF V600E mutations, but no other investigated mutation, and MSI-H are more commonly seen in an unselected cohort than is reported from clinical patient cohorts, likely because they indicate poor prognosis. Half of the patients have a tumor that is treatable with an already FDA-approved targeted drug for mCRC.
- 10 January 2020
- Published by Medical Journals Sweden AB in Acta Oncologica
- Vol. 59 (4) , 417-426
Open AccessMolecular characterization of a large unselected cohort of metastatic colorectal cancers in relation to primary tumor location, rare metastatic sites and prognosis
Background: We have reported that BRAF V600E mutations and microsatellite instability-high (MSI-H) are more prevalent in a population-based cohort of metastatic colorectal cancer (mCRC) patients than has been reported from clinical trials or hospital-based patient groups. The aim was to explore if other mutations in mCRC differ in prevalence between these cohorts in relation to mismatch repair status and primary tumor location and if presence of bone or brain metastases is associated with any mutations. Material and methods: A population-based cohort of 798 mCRC patients from three regions in Scandinavia was used. Forty-four cancer related genes were investigated in a custom designed Ampliseq hotspot panel. Differences in survival were analyzed using the Kaplan–Meier estimator and the Cox regression analysis. Results: Determination of mutations was possible in 449/501 patients for 40/44 genes. Besides BRAF V600E, seen in 19% of the tumors, none of the other mutations appeared more prevalent than in trial cohorts. BRAF V600E and MSI-H, seen in 8%, were associated with poor prognosis as was right-sided primary tumor location (39%) when compared to left-sided and rectum together; however, in a multivariable regression, only the BRAF mutation retained its statistical significance. No other mutations were associated with poor prognosis. ERBB2 alterations were more common if bone metastases were present at diagnosis (17% vs. 4%, p = .011). No association was found for brain metastases. Fifty-two percent had an alteration that is treatable with an FDA-approved targeted therapy, chiefly by EGFR-inhibitor for RAS wild-type and a check-point inhibitor for MSI-H tumors. Conclusions: Right-sided tumor location, BRAF V600E mutations, but no other investigated mutation, and MSI-H are more commonly seen in an unselected cohort than is reported from clinical patient cohorts, likely because they indicate poor prognosis. Half of the patients have a tumor that is treatable with an already FDA-approved targeted drug for mCRC.
- Open Access
- 26 November 2019
- Published by Medical Journals Sweden AB in Acta Oncologica
- Vol. 59 (3) , 284-290
Metastatic colorectal carcinomas with high SATB2 expression are associated with better prognosis and response to chemotherapy: a population-based Scandinavian study
Background: Survival and response to therapy in patients with metastatic colorectal cancer (mCRC) are very heterogeneous. There is an unmet need for better markers of prognosis and treatment benefit for mCRC patients. The homeobox 2 gene SATB2 has a highly specific expression in colorectal tissue and is associated with better prognosis in non-metastatic CRC. Material and methods: A population-based cohort of 798 mCRC patients was analysed. From primary tumour material, protein expression was assessed by immunohistochemistry. BRAF and KRAS mutation status was also determined. Associations with clinicopathological data, overall and progression-free survival and response to first-line chemotherapy were analysed. Results: Tumour tissue and clinical data were available from 467 patients. SATB2 was strongly expressed in 58% of cases, significantly more in left-sided, low-grade and wild-type BRAF tumours. Patients with high SATB2 tumours had longer overall survival compared with low SATB2 tumours (median 13 vs 8 months respectively, p < .001). Chemotherapy was given to 282 patients (63%). Patients with high SATB2 tumours had longer OS (median 22 vs 15 months respectively, p = .001) and more often responded to chemotherapy than those with low SATB2 (objective response 43% vs 29%, p = .02; clinical response 83% vs 67%, p = .004). Progression-free survival on first-line irinotecan chemotherapy was longer in high SATB2 cases (median 8 vs 4 months respectively, p = .019). Patients with both low SATB2 expression and mutated BRAF (n = 69) had particularly poor survival compared to the rest (median 8 and 12 months respectively, p = .001). In multivariable analysis, the SATB2 findings were independent of known clinicopathological prognostic markers, including BRAF mutation status. Conclusion: Patients with mCRC expressing high level of SATB2 have better prognosis and response to chemotherapy than those with low SATB2 expression. Patients with both low SATB2 expression and mutated BRAF had particularly poor prognosis and could thus benefit from more aggressive therapies.
- 26 November 2019
- Published by Medical Journals Sweden AB in Acta Oncologica
- Vol. 59 (3) , 284-290
Open AccessMetastatic colorectal carcinomas with high SATB2 expression are associated with better prognosis and response to chemotherapy: a population-based Scandinavian study
Background: Survival and response to therapy in patients with metastatic colorectal cancer (mCRC) are very heterogeneous. There is an unmet need for better markers of prognosis and treatment benefit for mCRC patients. The homeobox 2 gene SATB2 has a highly specific expression in colorectal tissue and is associated with better prognosis in non-metastatic CRC. Material and methods: A population-based cohort of 798 mCRC patients was analysed. From primary tumour material, protein expression was assessed by immunohistochemistry. BRAF and KRAS mutation status was also determined. Associations with clinicopathological data, overall and progression-free survival and response to first-line chemotherapy were analysed. Results: Tumour tissue and clinical data were available from 467 patients. SATB2 was strongly expressed in 58% of cases, significantly more in left-sided, low-grade and wild-type BRAF tumours. Patients with high SATB2 tumours had longer overall survival compared with low SATB2 tumours (median 13 vs 8 months respectively, p < .001). Chemotherapy was given to 282 patients (63%). Patients with high SATB2 tumours had longer OS (median 22 vs 15 months respectively, p = .001) and more often responded to chemotherapy than those with low SATB2 (objective response 43% vs 29%, p = .02; clinical response 83% vs 67%, p = .004). Progression-free survival on first-line irinotecan chemotherapy was longer in high SATB2 cases (median 8 vs 4 months respectively, p = .019). Patients with both low SATB2 expression and mutated BRAF (n = 69) had particularly poor survival compared to the rest (median 8 and 12 months respectively, p = .001). In multivariable analysis, the SATB2 findings were independent of known clinicopathological prognostic markers, including BRAF mutation status. Conclusion: Patients with mCRC expressing high level of SATB2 have better prognosis and response to chemotherapy than those with low SATB2 expression. Patients with both low SATB2 expression and mutated BRAF had particularly poor prognosis and could thus benefit from more aggressive therapies.
- Open Access
- 5 July 2019
- Published by Frontiers Media SA in Frontiers in Bioengineering and Biotechnology
- Vol. 7, 125
Glandular Segmentation of Prostate Cancer: An Illustration of How the Choice of Histopathological Stain Is One Key to Success for Computational Pathology
Digital pathology offers the potential for computer-aided diagnosis, significantly reducing the pathologists' workload and paving the way for accurate prognostication with reduced inter-and intra-observer variations. But successful computer-based analysis requires careful tissue preparation and image acquisition to keep color and intensity variations to a minimum. While the human eye may recognize prostate glands with significant color and intensity variations, a computer algorithm may fail under such conditions. Since malignancy grading of prostate tissue according to Gleason or to the International Society of Urological Pathology (ISUP) grading system is based on architectural growth patterns of prostatic carcinoma, automatic methods must rely on accurate identification of the prostate glands. But due to poor color differentiation between stroma and epithelium from the common stain hematoxylin-eosin, no method is yet able to segment all types of glands, making automatic prognostication hard to attain. We address the effect of tissue preparation on glandular segmentation with an alternative stain, Picrosirius red-hematoxylin, which clearly delineates the stromal boundaries, and couple this stain with a color decomposition that removes intensity variation. In this paper we propose a segmentation algorithm that uses image analysis techniques based on mathematical morphology and that can successfully determine the glandular boundaries. Accurate determination of the stromal and glandular morphology enables the identification of the architectural pattern that determine the malignancy grade and classify each gland into its appropriate Gleason grade or ISUP Grade Group. Segmentation of prostate tissue with the new stain and decomposition method has been successfully tested on more than 11000 objects including well-formed glands (Gleason grade 3), cribriform and fine caliber glands (grade 4), and single cells (grade 5) glands.
- 5 July 2019
- Published by Frontiers Media SA in Frontiers in Bioengineering and Biotechnology
- Vol. 7, 125
Open AccessGlandular Segmentation of Prostate Cancer: An Illustration of How the Choice of Histopathological Stain Is One Key to Success for Computational Pathology
Digital pathology offers the potential for computer-aided diagnosis, significantly reducing the pathologists' workload and paving the way for accurate prognostication with reduced inter-and intra-observer variations. But successful computer-based analysis requires careful tissue preparation and image acquisition to keep color and intensity variations to a minimum. While the human eye may recognize prostate glands with significant color and intensity variations, a computer algorithm may fail under such conditions. Since malignancy grading of prostate tissue according to Gleason or to the International Society of Urological Pathology (ISUP) grading system is based on architectural growth patterns of prostatic carcinoma, automatic methods must rely on accurate identification of the prostate glands. But due to poor color differentiation between stroma and epithelium from the common stain hematoxylin-eosin, no method is yet able to segment all types of glands, making automatic prognostication hard to attain. We address the effect of tissue preparation on glandular segmentation with an alternative stain, Picrosirius red-hematoxylin, which clearly delineates the stromal boundaries, and couple this stain with a color decomposition that removes intensity variation. In this paper we propose a segmentation algorithm that uses image analysis techniques based on mathematical morphology and that can successfully determine the glandular boundaries. Accurate determination of the stromal and glandular morphology enables the identification of the architectural pattern that determine the malignancy grade and classify each gland into its appropriate Gleason grade or ISUP Grade Group. Segmentation of prostate tissue with the new stain and decomposition method has been successfully tested on more than 11000 objects including well-formed glands (Gleason grade 3), cribriform and fine caliber glands (grade 4), and single cells (grade 5) glands.
- Open Access
- 9 May 2019
- Published by Wiley in Cancer Medicine
- Vol. 8 (7) , 3623-3635
Consequences of a high incidence of microsatellite instability and BRAF‐mutated tumors: A population‐based cohort of metastatic colorectal cancer patients
Background: Immunotherapy for patients with microsatellite‐instable (MSI‐H) tumors or BRAF‐inhibitors combination treatment for BRAF‐mutated (mutBRAF) tumors in metastatic colorectal cancer (mCRC) is promising, but the frequency of these molecular changes in trial patients are low. Unselected population‐based studies of these molecular changes are warranted.Methods: A population‐based cohort of 798 mCRC patients in Scandinavia was studied. Patient and molecular tumor characteristics, overall survival (OS) and progression‐free survival (PFS) were estimated.Results: Here, 40/583 (7%) tumor samples were MSI‐H and 120/591 (20%) were mutBRAF; 87% of MSI‐H tumors were mutBRAF (non‐Lynch). Elderly (>75 years) had more often MSI‐H (10% vs 6%) and MSI‐H/mutBRAF (9% vs 4%) tumors. Response rate (5% vs 44%), PFS (4 vs 8 months), and OS (9 vs 18 months) after first‐line chemotherapy was all significantly lower in patients with MSI‐H compared to patients with microsatellite stable tumors. MSI‐H and mutBRAF were both independent poor prognostic predictors for OS (P = 0.049, P < 0.001) and PFS (P = 0.045, P = 0.005) after first‐line chemotherapy. Patients with MSI‐H tumors received less second‐line chemotherapy (15% vs 37%, P = 0.005).Conclusions: In unselected mCRC patients, MSI‐H and mutBRAF cases were more common than previously reported. Patients with MSI‐H tumors had worse survival, less benefit from chemotherapy, and they differed considerably from recent third‐line immunotherapy trial patients as they were older and most had mutBRAF tumor (non‐Lynch).
- 9 May 2019
- Published by Wiley in Cancer Medicine
- Vol. 8 (7) , 3623-3635
Open AccessConsequences of a high incidence of microsatellite instability and BRAF‐mutated tumors: A population‐based cohort of metastatic colorectal cancer patients
Background: Immunotherapy for patients with microsatellite‐instable (MSI‐H) tumors or BRAF‐inhibitors combination treatment for BRAF‐mutated (mutBRAF) tumors in metastatic colorectal cancer (mCRC) is promising, but the frequency of these molecular changes in trial patients are low. Unselected population‐based studies of these molecular changes are warranted.Methods: A population‐based cohort of 798 mCRC patients in Scandinavia was studied. Patient and molecular tumor characteristics, overall survival (OS) and progression‐free survival (PFS) were estimated.Results: Here, 40/583 (7%) tumor samples were MSI‐H and 120/591 (20%) were mutBRAF; 87% of MSI‐H tumors were mutBRAF (non‐Lynch). Elderly (>75 years) had more often MSI‐H (10% vs 6%) and MSI‐H/mutBRAF (9% vs 4%) tumors. Response rate (5% vs 44%), PFS (4 vs 8 months), and OS (9 vs 18 months) after first‐line chemotherapy was all significantly lower in patients with MSI‐H compared to patients with microsatellite stable tumors. MSI‐H and mutBRAF were both independent poor prognostic predictors for OS (P = 0.049, P < 0.001) and PFS (P = 0.045, P = 0.005) after first‐line chemotherapy. Patients with MSI‐H tumors received less second‐line chemotherapy (15% vs 37%, P = 0.005).Conclusions: In unselected mCRC patients, MSI‐H and mutBRAF cases were more common than previously reported. Patients with MSI‐H tumors had worse survival, less benefit from chemotherapy, and they differed considerably from recent third‐line immunotherapy trial patients as they were older and most had mutBRAF tumor (non‐Lynch).
- Open Access
- 1 January 2019
- Published by Ivyspring International Publisher in Journal of Cancer
- Vol. 10 (14) , 3224-3231
Tumoral Pyruvate Kinase L/R as a Predictive Marker for the Treatment of Renal Cancer Patients with Sunitinib and Sorafenib
Background and aims: Treatment with tyrosine kinase inhibitors (TKI) like sunitinib and sorafenib has improved the prognosis of patients with metastatic renal cell cancer (mRCC). No predictive marker is available to select patients who will gain from these treatments. Tumoral pyruvate kinase L/R (PKLR) is a membrane protein with highly specific expression in the renal tubule. We have previously shown that the tumoral expression of cubilin (CUBN) is associated with progression free survival (PFS) in mRCC patients treated with sunitinib and sorafenib. The aim of the present study was to investigate if PKLR can predict response in these patients, alone and/or in combination with CUBN. Methods: A tissue microarray (TMA) was constructed of tumor samples from 139 mRCC patients. One hundred and thirty-six of these patients had been treated with sunitinib or sorafenib in the first or second-line setting. Thirty patients suffered from early severe toxicity leading to the termination of treatment. The remaining patients (n=106) were selected for the current study. Results: Fifty-five (52%) of the tumors expressed membranous PKLR. Patients with PKLR tumor expression experienced a significantly longer PFS compared to patients with no expression (eight versus five months, p = 0.019). Overall survival (OS) was also significantly better for patients with PKLR expression. In addition, the combined expression of PKLR and CUBN resulted in a higher predictive value than either marker alone. Conclusions: In this real world study we show that tumoral PKLR membrane expression is a positive predictive biomarker for sunitinib and sorafenib treatment in patients suffering from mRCC. Our results also indicate that the combined expression with cubilin more accurately than PKLR alone can select patients with no benefit from treatment.
- 1 January 2019
- Published by Ivyspring International Publisher in Journal of Cancer
- Vol. 10 (14) , 3224-3231
Open AccessTumoral Pyruvate Kinase L/R as a Predictive Marker for the Treatment of Renal Cancer Patients with Sunitinib and Sorafenib
Background and aims: Treatment with tyrosine kinase inhibitors (TKI) like sunitinib and sorafenib has improved the prognosis of patients with metastatic renal cell cancer (mRCC). No predictive marker is available to select patients who will gain from these treatments. Tumoral pyruvate kinase L/R (PKLR) is a membrane protein with highly specific expression in the renal tubule. We have previously shown that the tumoral expression of cubilin (CUBN) is associated with progression free survival (PFS) in mRCC patients treated with sunitinib and sorafenib. The aim of the present study was to investigate if PKLR can predict response in these patients, alone and/or in combination with CUBN. Methods: A tissue microarray (TMA) was constructed of tumor samples from 139 mRCC patients. One hundred and thirty-six of these patients had been treated with sunitinib or sorafenib in the first or second-line setting. Thirty patients suffered from early severe toxicity leading to the termination of treatment. The remaining patients (n=106) were selected for the current study. Results: Fifty-five (52%) of the tumors expressed membranous PKLR. Patients with PKLR tumor expression experienced a significantly longer PFS compared to patients with no expression (eight versus five months, p = 0.019). Overall survival (OS) was also significantly better for patients with PKLR expression. In addition, the combined expression of PKLR and CUBN resulted in a higher predictive value than either marker alone. Conclusions: In this real world study we show that tumoral PKLR membrane expression is a positive predictive biomarker for sunitinib and sorafenib treatment in patients suffering from mRCC. Our results also indicate that the combined expression with cubilin more accurately than PKLR alone can select patients with no benefit from treatment.