Validation of a prediction model for avoiding post-chemotherapy retroperitoneal lymphadenectomy in patients with metastatic nonseminomatous germ cell cancer

Punjani, Nahid; Power, Nicholas; Vanhie, James J; Winquist, Eric

doi:10.5489/cuaj.3558

Cited by 8 publications

(5 citation statements)

References 16 publications

(40 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…7 Despite a high discriminative accuracy (AUC range, 0.77 to 0.84), 7 this model has not been universally adopted in clinical practice because of its complexity. 20 With the use of radiomics, the current study identifies an AUC of 0.74 that improved to 0.80 in residual masses < 20 mm with the addition of clinical variables. Of note, the performance of our radiomics signature improved as the axial diameters increased, with an AUC of 0.58 at 40-mm axial cuts and 0.74 with unrestricted size criteria.…”

Section: Discussionmentioning

confidence: 75%

Applying Radiomics to Predict Pathology of Postchemotherapy Retroperitoneal Nodal Masses in Germ Cell Tumors

Lewin

Dufort

Halankar

et al. 2018

JCO Clinical Cancer Informatics

View full text Add to dashboard Cite

PurposeAfter chemotherapy, approximately 50% of patients with metastatic testicular germ cell tumors (GCTs) who undergo retroperitoneal lymph node dissections (RPNLDs) for residual masses have fibrosis. Radiomics uses image processing techniques to extract quantitative textures/features from regions of interest (ROIs) to train a classifier that predicts outcomes. We hypothesized that radiomics would identify patients with a high likelihood of fibrosis who may avoid RPLND.Patients and MethodsPatients with GCT who had an RPLND for nodal masses > 1 cm after first-line platinum chemotherapy were included. Preoperative contrast-enhanced axial computed tomography images of retroperitoneal ROIs were manually contoured. Radiomics features (n = 153) were used to train a radial basis function support vector machine classifier to discriminate between viable GCT/mature teratoma versus fibrosis. A nested 10-fold cross-validation protocol was used to determine classifier accuracy. Clinical variables/restricted size criteria were used to optimize the classifier.ResultsSeventy-seven patients with 102 ROIs were analyzed (GCT, 21; teratoma, 41; fibrosis, 40). The discriminative accuracy of radiomics to identify GCT/teratoma versus fibrosis was 72 ± 2.2% (area under the curve [AUC], 0.74 ± 0.028); sensitivity was 56.2 ± 15.0%, and specificity was 81.9 ± 9.0% (P = .001). No major predictive differences were identified when data were restricted by varying maximal axial diameters (AUC range, 0.58 ± 0.05 to 0.74 ± 0.03). The prediction algorithm using clinical variables alone identified an AUC of 0.76. When these variables were added to the radiomics signature, the best performing classifier was identified when axial masses were limited to diameter < 2 cm (accuracy, 88.2 ± 4.4; AUC, 0.80 ± 0.05; P = .02).ConclusionA predictive radiomics algorithm had a discriminative accuracy of 72% that improved to 88% when combined with clinical predictors. Additional independent validation is required to assess whether radiomics allows patients with a high predicted likelihood of fibrosis to avoid RPLND.

show abstract

Section: Discussionmentioning

confidence: 75%

Applying Radiomics to Predict Pathology of Postchemotherapy Retroperitoneal Nodal Masses in Germ Cell Tumors

Lewin

Dufort

Halankar

et al. 2018

JCO Clinical Cancer Informatics

View full text Add to dashboard Cite

show abstract

“…Almost half of the patients that undergo the surgery do not benefit from the procedure. In this study we used a population-based dataset, RETROP, to validate the prediction model used by two independent centres [13,14]. However, the study by Punjani et Vergouwe's prediction model shows good reproducibility and discrimination when applied to the RETROP data and hence good validity, despite the 30 years that have passed since the first patient was included in the Vergouwe study.…”

Section: Discussionmentioning

confidence: 99%

“…for predicting benign disease at PC‐RPLND. The Vergouwe model has previously been externally validated by two independent centres [13,14]. However, the study by Punjani et al.…”

Section: Discussionmentioning

confidence: 99%

“…The early version of the prediction model has been validated by the Vergouwe group [10][11][12]. Since the final version was published in 2007, two groups, one from Canada and one from Germany, have independently validated the prediction model [13,14]. However, the implementation of the model in clinical practice is challenging because of the risk of leaving teratoma or viable cancer in residual retroperitoneal lymph nodes.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Validation of a prediction model for post‐chemotherapy fibrosis in nonseminoma patients

et al. 2023

View full text Add to dashboard Cite

ObjectiveTo validate Vergouwe's prediction model using the Swedish and Norwegian Testicular Cancer Group (SWENOTECA) RETROP database and to define its clinical utility.Materials and methodsVergouwe's prediction model for benign histopathology in post‐chemotherapy retroperitoneal lymph node dissection (PC‐RPLND) uses the following variables: presence of teratoma in orchiectomy specimen; pre‐chemotherapy level of alpha‐fetoprotein; β‐Human chorionic gonadotropin and lactate dehydrogenase; and lymph node size pre‐ and post‐chemotherapy. Our validation cohort consisted of patients included in RETROP, a prospective population‐based database of patients in Sweden and Norway with metastatic nonseminoma, who underwent PC‐RPLND in the period 2007–2014. Discrimination and calibration analyses were used to validate Vergouwe's prediction model results. Calibration plots were created and a Hosmer–Lemeshow test was calculated. Clinical utility, expressed as opt‐out net benefit (NBopt‐out), was analysed using decision curve analysis.ResultsOverall, 284 patients were included in the analysis, of whom 130 (46%) had benign histology after PC‐RPLND. Discrimination analysis showed good reproducibility, with an area under the receiver‐operating characteristic curve (AUC) of 0.82 (95% confidence interval 0.77–0.87) compared to Vergouwe's prediction model (AUC between 0.77 and 0.84). Calibration was acceptable with no recalibration. Using a prediction threshold of 70% for benign histopathology, NBopt‐out was 0.098. Using the model and this threshold, 61 patients would have been spared surgery. However, only 51 of 61 were correctly classified as benign.ConclusionsThe model was externally validated with good reproducibility. In a clinical setting, the model may identify patients with a high chance of benign histopathology, thereby sparing patients of surgery. However, meticulous follow‐up is required.

show abstract

“…Therefore, this potentially very complex and morbid procedure [18] could be avoided in approximately half of these patients if we could predict which residual masses will only contain necrosis [19]. Towards this goal, predictive models have become available [20, 21] but their clinical use remains limited by low accuracy [22], although this may be improved by minor modifications [23]. Furthermore, a recent retrospective analysis suggested that disease recurrence can occur, albeit infrequently, even in cases where only residual fibrosis/necrosis is found following RPLND [24].…”

Section: Strategies To Minimize Treatment-related Morbiditymentioning

confidence: 99%

Recent developments in the management of germ cell tumors

Msaouel

Bilen

Zhang³

et al. 2017

Current Opinion in Oncology

View full text Add to dashboard Cite

Purpose of review In the present review, we summarize the recent developments in the management of germ cell tumors (GCTs). Recent findings Treatment-related acute and late-onset toxicity remains a key challenge in the management of GCTs. Recent data show that patients with large retroperitoneal lymph node metastases are at increased risk of venous thromboembolism and may benefit from prophylactic anticoagulation. Predictive models have been developed to identify patients with residual retroperitoneal lymph node masses that are more likely to benefit from surgical resection. However, their clinical use remains hampered by relatively low accuracy. There are currently multiple conventional-dose chemotherapy (CDCT) options for salvage therapy in patients with refractory or recurrent disease. In addition, more efficacious high-dose chemotherapy (HDCT) regimens continue to be developed. The role of salvage CDCT versus HDCT is currently being prospectively investigated. Finally, intratumoral heterogeneity is a common finding in cancer and an obvious observation in GCTs. Despite intratumoral heterogeneity, recent studies on nonseminomatous GCT have identified distinct histological subgroups and a potentially lethal clinical phenotype. Importantly, comprehensive molecular profiling so far has not elucidated the biologic basis or the clinical underpinnings of intratumoral heterogeneity in GCTs. Summary Remaining challenges to be addressed include minimizing therapeutic toxicity, and improving outcomes in patients with refractory/recurrent GCTs or malignant transformation of teratomas.

show abstract

Validation of a prediction model for avoiding post-chemotherapy retroperitoneal lymphadenectomy in patients with metastatic nonseminomatous germ cell cancer

Cited by 8 publications

References 16 publications

Applying Radiomics to Predict Pathology of Postchemotherapy Retroperitoneal Nodal Masses in Germ Cell Tumors

Applying Radiomics to Predict Pathology of Postchemotherapy Retroperitoneal Nodal Masses in Germ Cell Tumors

Validation of a prediction model for post‐chemotherapy fibrosis in nonseminoma patients

Recent developments in the management of germ cell tumors

Contact Info

Product

Resources

About