Predicting complete cytoreduction for advanced ovarian cancer patients using nearest-neighbor models

Laios, Alexandros; Gryparis, Alexandros; Jong, Diederick de; Hutson, Richard; Theophilou, Georgios; Leach, Chris

doi:10.1186/s13048-020-00700-0

Cited by 39 publications

(43 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The CT model of this study yielded an accuracy of 69.4% and an ROC area of 0.635, indicating a poor discrimination ability for differentiating the result of cytoreduction status. Similar results have been proposed by Liaos et al [ 23 ], with an accuracy of 66% for predicting complete cytoreduction of advanced ovarian cancer patients using nearest-neighbor models with attributes including clinical data and CT findings. Other studies showed a low sensitivity (27%) but high specificity (91%) for predicting incomplete or suboptimal cytoreduction using CT-PCI.…”

Section: Discussionsupporting

confidence: 82%

Adding Value of MRI over CT in Predicting Peritoneal Cancer Index and Completeness of Cytoreduction

Lin

Huang

et al. 2021

Diagnostics

View full text Add to dashboard Cite

Background: This study aimed to investigate the adding value of MRI over CT for preoperative cytoreductive surgery with hyperthermic intraperitoneal chemotherapies (CRS/HIPEC). Methods: Imaging and intraoperative peritoneal cancer index (PCI) were calculated in 62 patients with peritoneal metastasis. Predictive models for the completeness of cytoreductive score using PCI data were established using decision tree algorithms. Results: In gastric cancer patients, a large discrepancy and poor agreement was appreciated between CT and surgical PCI, and a nonsignificant difference was noted between MRI and surgical PCI. In colon cancer patients, a better agreement and higher correlation with a smaller error was observed in PCI score using MRI than in that using CT. However, the addition of MRI to CT was limited for appendiceal and ovarian cancer patients. For predicting incomplete cytoreduction, CT models yielded inadequate accuracy while MRI models were more accurate with fair discrimination ability. Conclusions: CT was suitable for estimating PCI and surgery outcome in appendiceal and ovarian cancer patients, while further MRI in addition to CT was recommended for colon and gastric cancer patients. However, for classifying patients with peritoneal carcinomatosis into complete and incomplete cytoreduction, MRI was more effective than CT.

show abstract

Section: Discussionsupporting

confidence: 82%

Adding Value of MRI over CT in Predicting Peritoneal Cancer Index and Completeness of Cytoreduction

Lin

Huang

et al. 2021

Diagnostics

View full text Add to dashboard Cite

show abstract

“…Developing methods to predict surgical outcomes is important to identify who will benefit from maximal cytoreductive effort in the primary or interval surgical setting. Data mining technologies appear to be promising for non-invasive, clinically meaningful improvements in the prediction accuracy of pre-surgical patient selection [35]. Although this approach may not be yet validated in ovarian cancer, it has the potential to be more reflective than other intra-operative assessments such as the PCI index.…”

Section: Discussionmentioning

confidence: 99%

The Uncertain Benefit of Adjuvant Chemotherapy in Advanced Low-Grade Serous Ovarian Cancer and the Pivotal Role of Surgical Cytoreduction

Johnson

Laios

Jackson

et al. 2021

JCM

Self Cite

View full text Add to dashboard Cite

In our center, adjuvant chemotherapy is routinely offered in high-grade serous ovarian cancer (HGSOC) patients but less commonly as a standard treatment in low-grade serous ovarian cancer (LGSOC) patients. This study evaluates the efficacy of this paradigm by analysing survival outcomes and by comparing the influence of different clinical and surgical characteristics between women with advanced LGSOC (n = 37) and advanced HGSOC (n = 300). Multivariate analysis was used to identify independent prognostic features for survival in LGSOC and HGSOC. Adjuvant chemotherapy was given in 99.7% of HGSOC patients versus in 27% of LGSOC (p < 0.0001). The LGSOC patients had greater surgical complexity scores (p < 0.0001), more frequent postoperative ICU/HDU admissions (p = 0.0002), and higher peri-/post-operative morbidity (p < 0.0001) compared to the HGSOC patients. The 5-year OS and progression-free survival (PFS) was 30% and 13% for HGSOC versus 57% and 21.6% for LGSOC, p = 0.016 and p = 0.044, respectively. Surgical complexity (HR 5.3, 95%CI 1.2–22.8, p = 0.024) and complete cytoreduction (HR 62.4, 95% CI 6.8–567.9, p < 0.001) were independent prognostic features for OS in LGSOC. This study demonstrates no clear significant survival advantage of chemotherapy in LGSOC. It highlights the substantial survival benefit of dynamic multi-visceral surgery to achieve complete cytoreduction as the primary treatment for LGSOC patients.

show abstract

“… 10 We previously demonstrated the feasibility of using a ML approach, the k-NN model, which is very much reflective of ‘previous clinical experience’ for accurate prediction of complete cytoreduction in advanced-stage HGSOC surgery. 11 …”

Section: Introductionmentioning

confidence: 99%

Feature Selection is Critical for 2-Year Prognosis in Advanced Stage High Grade Serous Ovarian Cancer by Using Machine Learning

et al. 2021

Self Cite

View full text Add to dashboard Cite

Introduction Accurate prediction of patient prognosis can be especially useful for the selection of best treatment protocols. Machine Learning can serve this purpose by making predictions based upon generalizable clinical patterns embedded within learning datasets. We designed a study to support the feature selection for the 2-year prognostic period and compared the performance of several Machine Learning prediction algorithms for accurate 2-year prognosis estimation in advanced-stage high grade serous ovarian cancer (HGSOC) patients. Methods The prognosis estimation was formulated as a binary classification problem. Dataset was split into training and test cohorts with repeated random sampling until there was no significant difference (p = 0.20) between the two cohorts. A ten-fold cross-validation was applied. Various state-of-the-art supervised classifiers were used. For feature selection, in addition to the exhaustive search for the best combination of features, we used the-chi square test of independence and the MRMR method. Results Two hundred nine patients were identified. The model's mean prediction accuracy reached 73%. We demonstrated that Support-Vector-Machine and Ensemble Subspace Discriminant algorithms outperformed Logistic Regression in accuracy indices. The probability of achieving a cancer-free state was maximised with a combination of primary cytoreduction, good performance status and maximal surgical effort (AUC 0.63). Standard chemotherapy, performance status, tumour load and residual disease were consistently predictive of the mid-term overall survival (AUC 0.63–0.66). The model recall and precision were greater than 80%. Conclusion Machine Learning appears to be promising for accurate prognosis estimation. Appropriate feature selection is required when building an HGSOC model for 2-year prognosis prediction. We provide evidence as to what combination of prognosticators leads to the largest impact on the HGSOC 2-year prognosis.

show abstract

Predicting complete cytoreduction for advanced ovarian cancer patients using nearest-neighbor models

Cited by 39 publications

References 16 publications

Adding Value of MRI over CT in Predicting Peritoneal Cancer Index and Completeness of Cytoreduction

Adding Value of MRI over CT in Predicting Peritoneal Cancer Index and Completeness of Cytoreduction

The Uncertain Benefit of Adjuvant Chemotherapy in Advanced Low-Grade Serous Ovarian Cancer and the Pivotal Role of Surgical Cytoreduction

Feature Selection is Critical for 2-Year Prognosis in Advanced Stage High Grade Serous Ovarian Cancer by Using Machine Learning

Contact Info

Product

Resources

About