Kari S. Wagner-Larsen’s research while affiliated with University of Bergen and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (21)


Fig. 2 MRI in a 30-year-old woman presenting with a large cervical tumor (squamous cell carcinoma, FIGO (2018) stage IIIC1) prior to treatment with concomitant radio-and chemotherapy. This patient had low myometrium ADC /tumor ADCmean -ratio at primary diagnosis and she had no signs of recurrence after 8.5 years. Coronal (a) and sagittal (b) T2-weighted MRI depicts a large cervical tumor (green arrows; with maximum diameter 8.5 cm) and disrupted stromal ring (white arrows) (a,b), tumor growth into the upper 2/3 of the vagina (blue arrows) (b) and enlarged (short axis diameter > 1 cm) iliac lymph nodes (open white arrows) (a). Axial oblique (relative to the long axis of the cervix) ADC-maps (c-f) depict restricted diffusion in the primary tumor. The following regions of interest were drawn on the ADC maps: tumorADC 1-5 , tumor ADCwhole bladder ADC , cervix ADC and myometrium ADC. Tumor ADCmean was calculated as the mean of tumor ADC1-5 , derived by two independent readers
Fig. 3 MRI in a 66-year-old woman presenting with a moderately large cervical tumor (squamous cell carcinoma, FIGO (2018) stage IIIC1) prior to treatment with concomitant radio-and chemotherapy. This patient had high myometrium ADC /tumor ADCmean -ratio and eventually died from cervical cancer 4.5 years after primary diagnosis. Axial oblique (relative to the long axis of the cervix) (a) and sagittal (b) T2-weighted MRI depicts an infiltrative moderately large (maximum tumor diameter of 3.9 cm) tumor (green arrows; a,b)in the uterine cervix with disrupted stromal ring to the right indicating parametrial invasion (white arrows; a), tumor growth into the upper 2/3 of the vagina (blue arrows; b) and enlarged right-sided iliac lymph node (white open arrows; a). Para-axial ADC-maps (c-f) depict restricted diffusion in the primary tumor. Regions of interest were drawn on the the ADC maps: tumor ADC1-5 , tumor ADCwhole bladder ADC , cervix ADC and myometrium ADC and tumor ADCmean was calculated as the mean of tumor ADC1-5 , derived by two independent readers
Fig. 4 Time-dependent receiver operating characteristic (tdROC) analysis with iAUC at 5 years (a,c) and AUC at 3 years (b,d) after diagnosis for predicting disease-specific survival (DSS) in uterine cervical cancer (CC). Normalizing tumor ADCmean to myometrium ADC by calculating a myometrium ADC /tumor ADCmean ratio yielded higher iAUC than for tumor ADCwhole (iAUC: 0.68 vs.0.59, P = 0.006), and tended to yield higher iAUC than for tumor ADCmean (iAUC: 0.68 vs.0.64, P = 0.09) (a). The tdROC-AUC at 3 years was higher for myometrium ADC /tumor ADCmean than for tumor ADCwhole and tumor ADCmean (AUC at 3 years: 0.71 vs. 0.57 and 0.64, respectively)(b). Myometrium ADC /tumor ADCmean , bladder ADC /tumor ADCmean and cervix ADC /tumor ADCmean yielded similar iAUCs and AUCs (P ≥ 0.12). The myometrium ADC /tumor ADCmean combined with FIGO (2018) yielded higher discriminatory performance for predicting DSS than FIGO (2018) alone (iAUC: 0.82 vs. 0.78, P = 0.02; AUC: 0.87 vs.0.82, P = 0.007) (c, d). ADC, apparent diffusion coefficient (10 −6 mm 2 /sec); FIGO, International Federation of Gynecology and Obstetrics; iAUC, the integrated area under the curve for the specified time interval; tumor-ADC, ADC measurements from the primary tumor
Fig. 5 Receiver operating characteristics (ROC) curves displaying the discriminatory abilities for predicting disease-specific survival in cervical cancer within FIGO (2018) stage I (a), stage II (b), stage III (c) at 5 years. The ROC curves were used to calculate the Youden indexes/optimal cut-offs within FIGO (2018) stages I, II, and III for high/low-myometrium ADC /tumor ADCmean groups. The high-myometrium ADC / tumor ADCmean groups had significantly lower survival than the low-myometrium ADC / tumor ADCmean group within FIGO (2018) stages I (d), stage II (e) and stage III (f). P-values were derived using the Log Rank test to compare survival distributions
Tumor ADC value predicts outcome and yields refined prognostication in uterine cervical cancer
  • Article
  • Full-text available

February 2025

·

16 Reads

Cancer Imaging

·

Kari S. Wagner-Larsen

·

Stian Ryste

·

[...]

·

Pelvic MRI is essential for evaluating local and regional tumor extent in uterine cervical cancer (CC). Tumor microstructure captured by diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) markers may be closely linked to prognosis in CC. Purpose To explore whether primary tumor ADC markers predict survival in CC. Material and methods CC patients ( n = 179) diagnosed during 2009–2020 with MRI-assessed primary maximum tumor size ≥ 2 cm were included in this retrospective single-center study. Two radiologists read all MRIs independently, measuring mean tumor ADC values in manually drawn regions of interest (ROIs) and mean tumor ADC (tumor ADCmean ) from five measurements for the two readers was used. ADC from ROIs in the myometrium (myometrium ADC ), cervical stroma (cervix ADC ), and bladder (bladder ADC ) were used to calculate ADC ratios. ADC markers were explored in relation to the International Federation of Gynecology and Obstetrics (FIGO) (2018) stage, disease-specific survival (DSS), and recurrence/progression-free survival (RPFS). Results Inter-reader agreement for all ADC measurements was high (ICC:0.59–0.79). Low tumor ADCmean predicted advanced FIGO stage ( P = 0.04) and reduced DSS (hazard ratio (HR): 0.96, P < 0.001; AIC: 441). Myometrium ADC /tumor ADCmean yielded the best Cox regression fit (AIC = 430) among all tumor ADC markers. Patients with high myometrium ADC /tumor ADCmean had significantly reduced 5-year DSS for FIGO stage I, II, and III ( P = 0.01, 0.004, and 0.02, respectively) and tended to the same for FIGO IV ( P = 0.22). Conclusion Low tumor ADCmean predicted reduced DSS in CC. High myometrium ADC /tumor ADCmean was the strongest ADC predictor of poor DSS and a marker of high-risk phenotype independent of FIGO stage.

Download

MRI delta radiomics during chemoradiotherapy for prognostication in locally advanced cervical cancer

January 2025

·

17 Reads

BMC Cancer

Background Effective diagnostic tools for prompt identification of high-risk locally advanced cervical cancer (LACC) patients are needed to facilitate early, individualized treatment. The aim of this work was to assess temporal changes in tumor radiomics (delta radiomics) from T2-weighted imaging (T2WI) during concurrent chemoradiotherapy (CCRT) in LACC patients, and their association with progression-free survival (PFS). Furthermore, to develop, validate, and compare delta- and pretreatment radiomic signatures for prognostic modeling. Methods A total of 110 LACC patients undergoing CCRT with MRI at baseline and mid-treatment were divided into training (cohortT: n = 73) and validation (cohortV: n = 37) cohorts. Radiomic features were extracted from tumors segmented on pre-CCRT and mid-CCRT T2WI and radiomic deltas (delta features) were computed. Two radiomic signatures for predicting PFS were constructed by least absolute shrinkage and selection operator (LASSO) Cox regression: Deltarad (from delta features) and Pre-CCRTrad (from pre-CCRT features). Prognostic performance of the radiomic signatures, 2018 International Federation of Gynecology and Obstetrics (FIGO) stage (I–IV), and baseline MRI-derived maximum tumor diameter (Tumormax: ≤2 cm; >2 and ≤ 4 cm; >4 cm) was evaluated by area under time-dependent receiver operating characteristics (tdROC) curves (AUC) in cohortT and cohortV (AUCT/AUCV). Mann–Whitney U tests assessed differences in radiomic delta features. PFS was evaluated using the Kaplan–Meier method with log-rank tests. Results Deltarad (AUCT/AUCV: 0.74/0.79) marginally outperformed Pre-CCRTrad (0.72/0.75) for predicting 5-year PFS, and both signatures clearly surpassed that of FIGO (0.61/0.61) and Tumormax (0.58/0.65). In total, four features within Deltarad and Pre-CCRTrad significantly differed in delta feature values between progressors and non-progressors, being consistently lower in progressors (p ≤ 0.03 for all). High Deltarad and Pre-CCRTrad radiomic scores were associated with poor PFS (p ≤ 0.04 for Deltarad in cohortT/Pre-CCRTrad in both cohorts; p = 0.11 for Deltarad in cohortV). Conclusions Delta- and pretreatment radiomic signatures equally allow early prognostication in LACC, outperforming FIGO stage and MRI-assessed maximum tumor diameter.


Impact of MRI radiomic feature normalization for prognostic modelling in uterine endometrial and cervical cancers

July 2024

·

55 Reads

·

2 Citations

Widespread clinical use of MRI radiomic tumor profiling for prognostication and treatment planning in cancers faces major obstacles due to limitations in standardization of radiomic features. The purpose of the current work was to assess the impact of different MRI scanning- and normalization protocols for the statistical analyses of tumor radiomic data in two patient cohorts with uterine endometrial-(EC) (n = 136) and cervical (CC) (n = 132) cancer. 1.5 T and 3 T, T1-weighted MRI 2 min post-contrast injection, T2-weighted turbo spin echo imaging, and diffusion-weighted imaging were acquired. Radiomic features were extracted from within manually segmented tumors in 3D and normalized either using z-score normalization or a linear regression model (LRM) accounting for linear dependencies with MRI acquisition parameters. Patients were clustered into two groups based on radiomic profile. Impact of MRI scanning parameters on cluster composition and prognostication were analyzed using Kruskal–Wallis tests, Kaplan–Meier plots, log-rank test, random survival forests and LASSO Cox regression with time-dependent area under curve (tdAUC) (α = 0.05). A large proportion of the radiomic features was statistically associated with MRI scanning protocol in both cohorts (EC: 162/385 [42%]; CC: 180/292 [62%]). A substantial number of EC (49/136 [36%]) and CC (50/132 [38%]) patients changed cluster when clustering was performed after z-score-versus LRM normalization. Prognostic modeling based on cluster groups yielded similar outputs for the two normalization methods in the EC/CC cohorts (log-rank test; z-score: p = 0.02/0.33; LRM: p = 0.01/0.45). Mean tdAUC for prognostic modeling of disease-specific survival (DSS) by the radiomic features in EC/CC was similar for the two normalization methods (random survival forests; z-score: mean tdAUC = 0.77/0.78; LRM: mean tdAUC = 0.80/0.75; LASSO Cox; z-score: mean tdAUC = 0.64/0.76; LRM: mean tdAUC = 0.76/0.75). Severe biases in tumor radiomics data due to MRI scanning parameters exist. Z-score normalization does not eliminate these biases, whereas LRM normalization effectively does. Still, radiomic cluster groups after z-score- and LRM normalization were similarly associated with DSS in EC and CC patients.


Exclusion criteria, overlapping data in study cohort and examples of segmented tumors on MRI. (a) The study cohort is established from a patient cohort of consenting patients (participation rate > 95%) with histopathologically confirmed cervical cancer diagnosed from 2009 to 2017 at Haukeland University Hospital (Bergen, Norway). The included patients were diagnosed with FIGO 2018 stage ≥ 1B1, had visible tumor on MRI, and an imaging protocol comprising axial (oblique) T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI). (b) Within the study cohort, 92, 65 and 73 patients had available biomarker, mutational and transcriptomic data, respectively. Clinicopathological data including extensive follow up were available for all 132 included patients. In total, 18 patients had no biomarker, mutational nor transcriptomic data available. Cervical cancer depicted on magnetic resonance imaging (MRI) by sagittal T2-weighted imaging (T2WI) (c,g), axial oblique (with manually segmented tumor mask) T2WI (d,h), and axial oblique/axial diffusion weighted imaging (DWI) (e,i) with corresponding apparent diffusion coefficient (ADC) maps (f,j) in two different patients allocated to Cluster 1 and Cluster 3, respectively. The tumor masks were drawn from T2WI supported by high b-value and ADC map. (c–f) (Cluster 1 patient): A 38-year-old woman diagnosed with a FIGO stage IIB (tumor invading the parametrium) squamous cell carcinoma with MRI assessed maximum tumor size of 4.7 cm receiving radio-chemotherapy as primary treatment; the patient was alive without signs of recurrence 5 years after diagnosis. (g–j) (Cluster 3 patient): A 67-year-old woman diagnosed with FIGO stage IIA2 (tumor invading upper two thirds of the vagina and size > 4.0 cm) squamous cell carcinoma with MRI assessed maximum tumor size of 4.2 cm receiving radio-chemotherapy as primary treatment; the patient experienced recurrence and died from disease 88 days after primary diagnosis.
Unsupervised clustering of 293 radiomic features in 132 cervical cancer patients yields three distinct patient clusters exhibiting significant different risk profiles. (a) Unsupervised k-medoid clustering of 293 radiomic features identified three patient clusters exhibiting distinct radiomic profiles. Each vertical line (values along x-axis) represents one radiomic feature, and each horizontal line (values along y-axis) represents one patient. The tumor volumes, mean apparent diffusion coefficient (ADC) values, FIGO stage and histologic types for the same patients are displayed in right panels. The radiomic features extracted from T2-weighted images, apparent diffusion coefficient (ADC) maps and high b-value diffusion weighted images (DWI) are indicated by arrows. Similarly, radiomic feature groups (firstorder, glcm, glszm, ngtdm, glrlm, gldm) are shown by unique color codes. (b,c) Significantly different disease-specific survival (DSS) for patients within the three radiomic clusters was observed in the entire patient cohort (b) and for the subgroup of patients with squamous cell carcinoma (c). P-values of DSS differences between specific clusters are indicated in gray. Kaplan–Meier curves depict probability values from Mantel-Cox log-rank test comparing categories. Number of patients/events for each category is given in parentheses. ADC: apparent diffusion coefficient, FIGO: International Federation of Gynaecology and Obstetrics, SCC: squamous cell carcinoma, AC: adenocarcinoma, glcm: gray level co-occurrence matrix, gldm: gray level dependence matrix, glrlm: gray level run length matrix, glszm: gray level size zone matrix, gtdm: neighboring gray tone difference matrix.
T2-derived radiomic features are most important for the radiomic clustering and extreme feature values associate with the high-risk cluster (Cluster 3). (a) Radiomic variables and their importance for the generated clusters, sorted according to descending feature ranking (FR) (from left to right). Higher Euclidean centroid distance (red dots) indicates larger separation of centroids and hence larger effect on the clustering. Orange, light blue and blue dots indicate centroid position of Cluster 1, 2 and 3, respectively, for each radiomic feature. Bar (b) indicates which MR series and (c) which feature group each radiomic feature is derived from. Two highest ranking feature intervals (shaded in pink) were recognized with characteristic profiles: (1) Features ranked from 1 to 8 have markedly elevated FRs and all features are derived from T2WI. (2) Features ranked 9–55, are also predominated by T2WI. (3) In both intervals, Cluster 3 (blue dots) associates with extreme (both positive and negative) centroid values. The number of features (n) derived from the respective MR series (b) and radiomic feature groups (c) is given in the plot. Tumor volume (plotted value highlighted in red) was ranked as 280 (out of the 293 features) for feature importance for the derived clustering (b). A detailed description of all 293 radiomic features and their contribution to the radiomic clustering is presented in Supplementary Note 1 and Supplementary Fig. 2, respectively. FR: feature ranking, glcm: gray level co-occurrence matrix, gldm: gray level dependence matrix, glrlm: gray level run length matrix, glszm: gray level size zone matrix, ngtdm: neighboring gray tone difference matrix.
Genomic, transcriptional, and molecular characterization of clusters reveal immunotherapy, CDK4/6 and YAP-TEAD inhibitors and p53 pathway plausible treatment strategies within clusters. (a) Oncoplot displaying the top 20 most frequently significantly mutated cervical cancer genes in relation to radiomic clusters. 16 of these genes (*) contain mutations that are druggable as defined by the Human Protein Atlas (https://www.proteinatlas.org/) (for more details, see Supplementary Table 7). (b) Proportion of patients with mutations within main oncogenic pathways in relation to radiomic cluster. (c) Distribution of L1000 log 2 expression levels of the 11 differentially expressed genes ((False Discovery Rate (FDR) < 0.01, Fold Change >  ± 1.75) relative to radiomic cluster. The distribution of gene expression values is significantly different between radiomic clusters (Kruskal–Wallis test; p ≤ 0.001). (d) Heatmap of protein expression for known prognostic biomarkers (p53, PD-L1, HLA-DQB1, LIMCH1) in cervical cancer cases ordered by, radiomic cluster, histological type, inflammatory reaction and FIGO2018 stage. Grey boxes indicate no biomarker data available. p53 aberrations (negative or strong staining) are more common in Cluster 3 tumors in 68% (17/27) vs. in 30% (18/61) of Cluster 1/2 tumors (γ2 test; p = 0.004). High LIMCH1 protein expression tended to be more common in Cluster 3 tumors (in 96% (25/26) vs. in 79% (48/61) of Cluster 1/2 tumors) (Fisher’s Exact Test; p = 0.056). AC: adenocarcinoma; ASC: adenosquamous carcinoma; CC: cervical cancer; NET: neuroendocrine tumor; SCC: squamous cell carcinoma; UDC: undifferentiated carcinoma.
Radiomic profiles improve prognostication and reveal targets for therapy in cervical cancer

May 2024

·

29 Reads

·

2 Citations

Cervical cancer (CC) is a major global health problem with 570,000 new cases and 266,000 deaths annually. Prognosis is poor for advanced stage disease, and few effective treatments exist. Preoperative diagnostic imaging is common in high-income countries and MRI measured tumor size routinely guides treatment allocation of cervical cancer patients. Recently, the role of MRI radiomics has been recognized. However, its potential to independently predict survival and treatment response requires further clarification. This retrospective cohort study demonstrates how non-invasive, preoperative, MRI radiomic profiling may improve prognostication and tailoring of treatments and follow-ups for cervical cancer patients. By unsupervised clustering based on 293 radiomic features from 132 patients, we identify three distinct clusters comprising patients with significantly different risk profiles, also when adjusting for FIGO stage and age. By linking their radiomic profiles to genomic alterations, we identify putative treatment targets for the different patient clusters (e.g., immunotherapy, CDK4/6 and YAP-TEAD inhibitors and p53 pathway targeting treatments).


Impact of MRI radiomic feature normalization for prognostic modelling in uterine endometrial and cervical cancers.

February 2024

·

30 Reads

Objectives Widespread clinical use of MRI radiomic tumor profiling for prognostication and treatment planning in cancers faces major obstacles due to limitations in standardization of radiomic features. The purpose of the current work was to assess the impact of different MRI scanning- and normalization protocols for the statistical analyses of tumor radiomic data in two patient cohorts with uterine endometrial- (EC) (n = 136) and cervical (CC) (n = 132) cancer. Material and methods 1.5 T and 3 T, T1-weighted MRI 2 minutes post-contrast injection, T2-weighted turbo spin echo imaging, and diffusion-weighted imaging were acquired. Radiomic features were extracted from within manually segmented tumors in 3D and normalized either using z-score normalization or a linear regression model (LRM) accounting for linear dependencies with MRI acquisition parameters. Patient clustering into two groups based on radiomic profile. Impact of MRI scanning parameters on cluster composition and prognostication by cluster groups were analyzed using Kruskal-Wallis tests, Kaplan-Meier plots, log-rank test and random survival forest time-dependent area under curve (tdAUC) (α = 0.05). Results A large proportion of the radiomic features was statistically associated with MRI scanning protocol in both cohorts (EC: 162/385 [42%]; CC: 180/292 [62%]). A substantial number of EC (49/136 [36%]) and CC (50/132 [38%]) patients changed cluster when clustering was performed after z-score- versus LRM normalization. Prognostic modeling based on cluster groups yielded similar outputs for the two normalization methods in the EC/CC cohorts (log-rank test; z-score: p = 0.02/0.33; LRM: p = 0.01/0.45). Mean tdAUC for prognostic modeling of disease-specific survival (DSS) by the radiomic features in EC/CC was similar for the two normalization methods (random survival forest; z-score: mean tdAUC = 0.77/0.78; LRM: mean tdAUC = 0.80/0.75). Conclusions Severe biases in tumor radiomics data due to MRI scanning parameters exist. Z-score normalization does not eliminate these biases, whereas LRM normalization effectively does. Still, radiomic cluster groups after z-score- and LRM normalization were associated with similar DSS in EC and CC patients.



MRI ‐based radiomic signatures for pretreatment prognostication in cervical cancer

October 2023

·

34 Reads

·

2 Citations

Background Accurate pretherapeutic prognostication is important for tailoring treatment in cervical cancer (CC). Purpose To investigate whether pretreatment MRI‐based radiomic signatures predict disease‐specific survival (DSS) in CC. Study Type Retrospective. Population CC patients ( n = 133) allocated into training (T) ( n T = 89)/validation (V) ( n V = 44) cohorts. Field Strength/Sequence T2‐weighted imaging (T2WI) and diffusion‐weighted imaging (DWI) at 1.5T or 3.0T. Assessment Radiomic features from segmented tumors were extracted from T2WI and DWI (high b ‐value DWI and apparent diffusion coefficient (ADC) maps). Statistical Tests Radiomic signatures for prediction of DSS from T2WI (T2 rad ) and T2WI with DWI (T2 + DWI rad ) were constructed by least absolute shrinkage and selection operator (LASSO) Cox regression. Area under time‐dependent receiver operating characteristics curves (AUC) were used to evaluate and compare the prognostic performance of the radiomic signatures, MRI‐derived maximum tumor size ≤/> 4 cm (MAX size ), and 2018 International Federation of Gynecology and Obstetrics (FIGO) stage (I–II/III–IV). Survival was analyzed using Cox model estimating hazard ratios (HR) and Kaplan–Meier method with log‐rank tests. Results The radiomic signatures T2 rad and T2 + DWI rad yielded AUC T /AUC V of 0.80/0.62 and 0.81/0.75, respectively, for predicting 5‐year DSS. Both signatures yielded better or equal prognostic performance to that of MAX size (AUC T /AUC V : 0.69/0.65) and FIGO (AUC T /AUC V : 0.77/0.64) and were significant predictors of DSS after adjusting for FIGO (HR T /HR V for T2 rad : 4.0/2.5 and T2 + DWI rad : 4.8/2.1). Adding T2 rad and T2 + DWI rad to FIGO significantly improved DSS prediction compared to FIGO alone in cohort (T) (AUC T 0.86 and 0.88 vs. 0.77), and FIGO with T2 + DWI rad tended to the same in cohort (V) (AUC V 0.75 vs. 0.64, p = 0.07). High radiomic score for T2 + DWI rad was significantly associated with reduced DSS in both cohorts. Data Conclusion Radiomic signatures from T2WI and T2WI with DWI may provide added value for pretreatment risk assessment and for guiding tailored treatment strategies in CC.


#870 Radiomic profiles improve prognostication and reveal targets for therapy in cervical cancer

September 2023

·

16 Reads

International Journal of Gynecological Cancer

Introduction/Background Pelvic magnetic resonance imaging (MRI) is an important part of primary diagnostic workup in cervical cancer (CC), with MRI-assessed tumour size and pelvic tumour extent routinely guiding treatment decisions. Extraction of MRI-derived radiomic tumour features could improve cancer prognostics and may also reveal novel targets for treatment. Methodology We manually segmented 3D volumes in 132 primary CC tumours and extracted 293 whole-volume radiomic MRI features. Unsupervised hierarchical clustering yielded three distinct patient clusters (Cluster 1 [n=52]; 2 [n=46]; and 3 [n=34]). Overlapping clinicopathologic, genomic (whole exome sequencing, n=65), transcriptomic (L1000 arrays, n=73) and molecular biomarker (n=84) data were utilized to characterize each cluster. Results Patients in Cluster 2 and 3 had significantly reduced disease-specific survival (DSS) (hazard rate [HR]: 3.33; p=0.008) compared with patients in Clusters 1, even after adjusting for International Federation of Gynecology and Obstetrics (FIGO) 2018 stage and age (adjusted HR: 2.51; p=0.045). Cluster 3 tumours associate with high stage (p<0.001), large tumours (p<0.001), squamous histology (p=0.015), p53 negative or -overexpressing tumours (p=0.04) and aberrant TP53, -MYC and -MTORC1 signalling. The intermediate-risk Cluster 2 associates with increased and aberrant cell cycle- and Hippo signalling, suggesting CDK2/4 and YAP-TEAD inhibitors as plausible treatment options. The low-risk Cluster 1 associates with increased immune cell signalling. Conclusion This study links radiomic signatures to distinct genomic profiles that may potentially aid in prognostication and tailoring of treatments and follow-up plans for cervical cancer patients. Disclosures The authors declare no conflict interests.


Visceral fat percentage for prediction of outcome in uterine cervical cancer

July 2023

·

18 Reads

·

5 Citations

Gynecologic Oncology

Objective: The prognostic role of adiposity in uterine cervical cancer (CC) is largely unknown. Abdominal fat distribution may better reflect obesity than body mass index. This study aims to describe computed tomography (CT)-assessed abdominal fat distribution in relation to clinicopathologic characteristics, survival, and tumor gene expression in CC. Methods: The study included 316 CC patients diagnosed during 2004-2017 who had pre-treatment abdominal CT. CT-based 3D segmentation of total-, subcutaneous- and visceral abdominal fat volumes (TAV, SAV and VAV) allowed for calculation of visceral fat percentage (VAV% = VAV/TAV). Liver density (LD) and waist circumference (at L3/L4-level) were also measured. Associations between CT-derived adiposity markers, clinicopathologic characteristics and disease-specific survival (DSS) were explored. Gene set enrichment of primary tumors were examined in relation to fat distribution in a subset of 108 CC patients. Results: High TAV, VAV and VAV% and low LD were associated with higher age (≥44 yrs.; p ≤ 0.017) and high International Federation of Gynecology and Obstetrics (FIGO) (2018) stage (p ≤ 0.01). High VAV% was the only CT-marker predicting high-grade histology (p = 0.028), large tumor size (p = 0.016) and poor DSS (HR 1.07, p < 0.001). Patients with high VAV% had CC tumors that exhibited increased inflammatory signaling (false discovery rate [FDR] < 5%). Conclusions: High VAV% is associated with high-risk clinical features and predicts reduced DSS in CC patients. Furthermore, patients with high VAV% had upregulated inflammatory tumor signaling, suggesting that the metabolic environment induced by visceral adiposity contributes to tumor progression in CC.


2022-RA-1383-ESGO High visceral fat percentage is linked to upregulated inflammatory tumour signalling and predicts poor outcome in uterine cervical cancer

October 2022

·

4 Reads

International Journal of Gynecological Cancer

Introduction/Background The aim of this study was to explore abdominal fat distribution markers from computed tomography (CT) in relation to clinicopathologic characteristics and patient outcome in uterine cervical cancer (CC). By unravelling possible links between fat distribution profiles and altered tumour signalling pathways, potential molecular targets for treatment based on body composition profiles may be identified, which may enable more individualized treatment strategies in CC. Methodology The study included 316 CC patients diagnosed during 2004–2017 who had pre-treatment abdominal CT scans. CT images were analysed to quantify total abdominal fat volume (TAV), subcutaneous abdominal fat volume (SAV), visceral abdominal fat volume (VAV), visceral fat percentage (VAV% = VAV/TAV x100), liver density (LD) and waist circumference (WC). CT morphometric markers were explored in relation to clinicopathologic characteristics and disease-specific survival (DSS), and to gene expression profiles (L1000 mRNA) in a subset of 108 patients. Results High TAV, VAV and VAV% and low LD were all associated with high (≥44 years) patient age (p≤0.017) and high International Federation of Gynaecology and Obstetrics (FIGO) (2018) stage (p≤0.01). High VAV% was the only CT marker predicting high-grade histology (p=0.028), large tumour size (p=0.016) and poor DSS (HR 1.06, p<0.001). VAV% was strongly positively correlated with age (r=0.68, p<0.001) and VAV (r=0.65, p<0.001). Patients with high VAV% had CC tumours with enrichment of gene sets (false discovery rate [FDR] <5%) related to inflammatory signalling with 65% (13/20) of the top ranked Gene Ontology gene sets related to interferon signalling, viral- or immune response. • Download figure • Open in new tab • Download powerpoint Abstract 2022-RA-1383-ESGO Figure 1 High visceral fat percentage is linked to upregulated inflammatory tumour signalling and predicts poor outcome in uterine cervical cancer. (A) Time-dependent receiver operating characteristic (tdROC) curves for predicting disease-specific survival (DSS) at 5 years after diagnosis based on visceral abdominal fat percentage (VAV%), visceral abdominal fat volume (VAV), total abdominal fat volume (TAV) and subcutaneous abdominal fat volume (SAV). VAV% yielded significantly higher AUC (0.75) than the other morphometric makers (P<0.001 for all). B) Kaplan-Meier plot depicting significantly reduced DSS in patients with VAV%≥29 compared with patients with VAV%<29 (p<0.001). C) Abdominal compared lomography (CT) scans with segmentation of visceral and subcutaneous fat compartments carcinoma, international federation of gynaecology and obstetrics (FIGO) (2018) stage III. Patient I, aged 61 yrs, who had low VAV% (23%) received primary radiation therapy and subsequent chemotherapy with cisplatin. She developed pelvic metastases and died from cervical cancer 14 months after primary treatment. D) Gene set enrichment analysis (GSEA) revealed that patients with VAV%>29 had tumours exhibiting upregulated signalling pathways for gene sets involved in inflammatory signalling and immune response (shown in green) Conclusion High VAV% is associated with high-risk clinical features and predicts reduced disease-specific survival in CC patients. CC patients with high VAV% have tumours with upregulated genes involved in inflammatory signalling, suggesting that the metabolic environment induced by visceral adiposity influences the regulatory signalling pathways relevant for tumour progression in CC.


Citations (13)


... However, their integration into routine clinical practice is still evolving. [13][14][15] European Society of Gynaecological Oncology/ European Society for Medical Oncology (ESMO-ESGO) guidelines also suggest that expert vaginal ultrasound examinations, performed by an expert sonographer, can be used for the detection of myometrial invasion and cervical stromal invasion instead of pelvic MRI. 16 The ability to distinguish between primary cervical and endometrial cancer is essential to formulate an effective treatment plan in the postoperative period. ...

Reference:

Challenges in differentiating between primary cervical cancer versus stage II endometrial cancer
Radiomic profiles improve prognostication and reveal targets for therapy in cervical cancer

... Adjuvant chemotherapy was administered to 52 patients, ensuring comprehensive treatment of their condition. According to the 2018 International Federation of Gynecology and Obstetrics (FIGO) staging criteria [15], the patient distribution across different stages was as follows: stage IB1 (n = 38), stage IB2 (n = 21), stage IIA1 (n = 16), and stage IIA2 (n = 2). Radiotherapy played a crucial role in the treatment plan, with external beam irradiation administered to 75 patients, while 2 individuals received external beam irradiation followed by brachytherapy. ...

Clinicopathological and radiological stratification within FIGO 2018 stages improves risk-prediction in cervical cancer
  • Citing Article
  • December 2023

Gynecologic Oncology

... Magnetic resonance imaging (MRI) serves as a crucial noninvasive tool for CC diagnosis and staging [12,13], offering detailed insights into tumor morphology and extent within the pelvis, including potential bladder and rectal invasion, and predicting responses to neoadjuvant chemotherapy [14,15]. Furthermore, radiomics models derived from MRI data can predict lymph node metastasis (LNM) and lymphovascular space invasion (LVSI), critical factors in determining post-operative care and patient outcomes [16,17]. ...

MRI ‐based radiomic signatures for pretreatment prognostication in cervical cancer

... Zexuan Hu et al. [21] evaluated visceral fat using CT scan and observed that visceral fat has high accuracy in predicting high-grade kidney cancer across different genders and is an effective independent predictor of high-grade kidney cancer in women. Park et al. [22] reported that Fuhrman classi cation of kidney cancer is signi cantly associated with the composition of visceral fat. In addition, the incidence of Fuhrman high-grade and high-stage tumors increases with the increase in the percentage of visceral fat tissue (VAT). ...

Visceral fat percentage for prediction of outcome in uterine cervical cancer
  • Citing Article
  • July 2023

Gynecologic Oncology

... In the diagnostic landscape of endometrial cancer, dynamic contrast-enhanced magnetic resonance imaging (MRI) can accurately assess myometrial invasion, cervical involvement, and lymph node (LN) metastasis [3,4]. Additionally, 18 F-fluorodeoxyglucose (FDG) positron emission tomography-computed tomography (PET-CT) serves as a valuable tool for detecting occult metastatic lesions [5]. Consequently, the utilization of both MRI and PET-CT in the preoperative radiologic assessment of endometrial cancer is on the rise, underscoring the increasing preference for these imaging modalities. ...

Preoperative pelvic MRI and 2-[ 18 F]FDG PET/CT for lymph node staging and prognostication in endometrial cancer-time to revisit current imaging guidelines?

European Radiology

... [9,14] and endometrial cancer (ICC: 0.60) [37]. Furthermore, the inter-reader agreement for tumor ADC measurements in the present study was comparable to the agreement reported for CC MRI-assessed maximum tumor diameter (ICC of 0.73 in patients with visible tumors [38]), a metric that is incorporated in the FIGO (2018) stage assignment. Altogether, the high inter-reader agreement for tumor ADC measurements supports its potential implementation as a prognostic marker in the clinic. ...

What MRI-based tumor size measurement is best for predicting long-term survival in uterine cervical cancer?

Insights into Imaging

... Recent advancements in deep learning (DL), including convolutional neural networks (CNNs), have been proven to have excellent performance in the general use of multiple detection and segmentation tasks in medical imaging [16][17][18][19][20], particularly for brain or cardiac diseases. Nevertheless, research on cervical cancer detection and segmentation has predominantly focused on singlesequence MRI or computed tomography (CT) images [8,[21][22][23][24][25][26][27], with relatively few systematic investigations of the segmentation performance across different sequences. Some studies have explored the analysis of multi-parametric MRI of cervical cancer but are primarily constrained by either limited sequences [28,29] or non-DL algorithms [30]. ...

Automatic Whole-Volume Tumor Segmentation in Cervical Cancer

... Consistency and reproducibility between readers are essential for all prognostic markers if they are to be introduced in the clinic [36]. In our study, we demonstrated high inter-reader agreement for measuring tumor ADC (ICC: 0.67-0.79), ...

Interobserver agreement and prognostic impact for MRI-based 2018 FIGO staging parameters in uterine cervical cancer

European Radiology

... Other studies have investigated radio-genomic correlations in EC 46,47 . Radiopathomic studies have been conducted in the context of other cancer types such as breast cancer, prostate cancer, and brain tumors 44,[48][49][50][51] .However, most of these studies focused on developing prediction models that combine features extracted from both radiological and digital pathology images, without exploring the underlying correlations among them. ...

A radiogenomics application for prognostic profiling of endometrial cancer

Communications Biology

... Previous studies have identified LIMCH1 as being both a tumour suppressor and an oncogene in different cancer types, including breast cancer, renal cancer, and lung adenocarcinoma 8,10,25 . Specifically, it has been recognized as being a tumour suppressor in lung cancer 7 , whereas in cervical and breast cancer, it has been implicated as an oncogene 4,11 . However, the precise role of LIMCH1 in tumorigenesis remains to be fully elucidated. ...

A 10-gene prognostic signature points to LIMCH1 and HLA-DQB1 as important players in aggressive cervical cancer disease

British Journal of Cancer