Predictive Nomogram for Early Recurrence after Pancreatectomy in Resectable Pancreatic Cancer: Risk Classification Using Preoperative Clinicopathologic Factors

Kim, Naru; Han, In Woong; Ryu, Yoshinori; Hwang, Dae Wook; Heo, Jin Seok; Choi, Dong Wook; Shin, Sang Hyun

doi:10.3390/cancers12010137

Cited by 35 publications

(42 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To the best of our knowledge, this is the first study to develop and compare machine learning-based models to predict relapse risk of pancreatic ductal adenocarcinoma after radical resection from multi-institutional datasets. Predictive nomograms based on conventional regression methods have been built for early recurrence after pancreatectomy in resectable pancreatic cancer [ 9 , 12 ]. Kim et al established a nomogram to predict the probability of recurrence within 12 months after surgery in single medical center with AUROC = 0.655 [ 9 ].…”

Section: Discussionmentioning

confidence: 99%

“…Predictive nomograms based on conventional regression methods have been built for early recurrence after pancreatectomy in resectable pancreatic cancer [ 9 , 12 ]. Kim et al established a nomogram to predict the probability of recurrence within 12 months after surgery in single medical center with AUROC = 0.655 [ 9 ]. While in our study, we constructed and externally validated a predictive SVM model for 1-year relapse risk with AUROC = 0.733 and a KNN model for 2-year relapse risk with AUROC = 0.689 using stringent statistical method.…”

Section: Discussionmentioning

confidence: 99%

“…Nowadays, development of methods to predict treatment outcomes and prognosis is an important paradigm in the realm of personalized medicine [ 8 ]. Several studies have shown comparable prediction accuracy by using traditional regression-based statistical methods on a basis of a combination of biomarkers and multiple clinical factors [ 9 – 12 ]. However, common statistical methods familiar to clinicians ignore more complex non-linear interactions between variables that might play significant roles in the potential of future relapse, and which could be captured using more sophisticated modeling approaches [ 13 ].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Multi-institutional development and external validation of machine learning-based models to predict relapse risk of pancreatic ductal adenocarcinoma after radical resection

Yang

Yuan³

et al. 2021

J Transl Med

View full text Add to dashboard Cite

Background Surgical resection is the only potentially curative treatment for pancreatic ductal adenocarcinoma (PDAC) and the survival of patients after radical resection is closely related to relapse. We aimed to develop models to predict the risk of relapse using machine learning methods based on multiple clinical parameters. Methods Data were collected and analysed of 262 PDAC patients who underwent radical resection at 3 institutions between 2013 and 2017, with 183 from one institution as a training set, 79 from the other 2 institution as a validation set. We developed and compared several predictive models to predict 1- and 2-year relapse risk using machine learning approaches. Results Machine learning techniques were superior to conventional regression-based analyses in predicting risk of relapse of PDAC after radical resection. Among them, the random forest (RF) outperformed other methods in the training set. The highest accuracy and area under the receiver operating characteristic curve (AUROC) for predicting 1-year relapse risk with RF were 78.4% and 0.834, respectively, and for 2-year relapse risk were 95.1% and 0.998. However, the support vector machine (SVM) model showed better performance than the others for predicting 1-year relapse risk in the validation set. And the k neighbor algorithm (KNN) model achieved the highest accuracy and AUROC for predicting 2-year relapse risk. Conclusions By machine learning, this study has developed and validated comprehensive models integrating clinicopathological characteristics to predict the relapse risk of PDAC after radical resection which will guide the development of personalized surveillance programs after surgery.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multi-institutional development and external validation of machine learning-based models to predict relapse risk of pancreatic ductal adenocarcinoma after radical resection

Yang

Yuan³

et al. 2021

J Transl Med

View full text Add to dashboard Cite

show abstract

“…The technique involves determining cancer DNA, vesicles and tumour cells in circulating blood [ 150 , 151 ]. Nomograms can be used to stratify patient risk for selection to treatment and may influence the choice of NAT, surgery or nature of adjuvant chemotherapy [ 152 , 153 , 154 , 155 , 156 ].…”

Section: Post-operative Pathwaysmentioning

confidence: 99%

Achieving ‘Marginal Gains’ to Optimise Outcomes in Resectable Pancreatic Cancer

Powell‐Brett

Pandé

Roberts

2021

Cancers

View full text Add to dashboard Cite

Improving outcomes among patients with resectable pancreatic cancer is one of the greatest challenges of modern medicine. Major improvements in survival will result from the development of novel therapies. However, optimising existing pathways, so that patients realise benefits of already proven treatments, presents a clear opportunity to improve outcomes in the short term. This narrative review will focus on treatments and interventions where there is a clear evidence base to improve outcomes in pancreatic cancer, and where there is also evidence of variation and under-treatment. Avoidance of preoperative biliary drainage, treatment of pancreatic exocrine insufficiency, prehabiliation and enhanced recovery after surgery, reducing perioperative complications, optimising opportunities for elderly patients to receive therapy, optimising adjuvant chemotherapy and regular surveillance after surgery are some of the strategies discussed. Each treatment or pathway change represents an opportunity for marginal gain. Accumulation of marginal gains can result in considerable benefit to patients. Given that these interventions already have evidence base, they can be realised quickly and economically.

show abstract

“…Recurrence within the first 12 months after surgery is considered to be an “early recurrence” (ER) and is a characteristic of a PDAC. The perioperative predictors of an ER such as the tumor size [ 6 , 7 , 8 ], metastases in the harvested lymph nodes [ 6 , 7 , 8 ], the serum carbohydrate antigen 19-9 (CA19-9) value [ 6 , 9 , 10 , 11 , 12 ], the duration of symptoms [ 13 ], a modified Glasgow Prognostic Score [ 14 ], a Charlson age-comorbidity index [ 6 ], tumor differentiation [ 6 , 12 , 13 ] and p53 expression in the primary tumor [ 15 ] have been reported to identify high-risk patients. Recently, neoadjuvant therapy has been shown to provide oncological benefits more than upfront surgery in patients with a borderline-resectable PDAC (BR-PDAC) [ 16 , 17 ]; significant survival benefits of neoadjuvant chemotherapy have also been demonstrated in phase III of the study on patients with a resectable PDAC (R-PDAC) [ 18 ].…”

Section: Introductionmentioning

confidence: 99%

Perioperative Predictors of Early Recurrence for Resectable and Borderline-Resectable Pancreatic Cancer

Imamura

Nagayama

Kyuno

et al. 2021

Cancers

View full text Add to dashboard Cite

We aimed to identify the perioperative predictors of the early recurrence (ER) of resectable and borderline-resectable pancreatic ductal adenocarcinomas (PDACs). After surgery for a PDAC, most patients develop a recurrence. Predictive factors may therefore guide therapeutic decision-making. Patients (n = 234) who underwent a pancreatectomy for a PDAC between 2006 and 2019 were included. The postrecurrence survival (PRS) was estimated using Kaplan–Meier curves. Predictive factors for an ER were assessed using logistic regression analyses; 93 patients (39.7%) were recurrence-free at the last follow-up. Patients with an ER (n = 85, 36.3%), defined as a recurrence within the first 12 months after surgery, had 1- and 2-year PRS rates of 38.7% and 9.5%, respectively, compared with 66.9% and 37.2% for those with a late recurrence (n = 56, 23.9%; both p < 0.001). The most common site of an ER was the liver (55.3%) with a significantly shorter median overall survival time than that with either a local or a lung recurrence (14.5 months; p < 0.001). Preoperative and postoperative risk factors for an ER included a tumor size >3.0 cm (odds ratio (OR): 3.11, 95% confidence interval (CI): 1.35–7.14) and preoperative carbohydrate antigen 19-9 (CA19-9) levels >52 U/mL (OR: 3.25, 95% CI: 1.67–6.30) and a pathological tumor size >3.0 cm (OR: 2.00, 95% CI: 1.03–3.90) and postoperative carbohydrate antigen 19-9 levels >37 U/mL (OR: 2.11, 95% CI: 1.02–4.36), respectively. Preoperatively (>52 U/mL) and postoperatively (>37 U/mL) elevated CA19-9 and a tumor size >3.0 cm were independent predictors for an ER after a pancreatectomy for a PDAC.

show abstract

Predictive Nomogram for Early Recurrence after Pancreatectomy in Resectable Pancreatic Cancer: Risk Classification Using Preoperative Clinicopathologic Factors

Cited by 35 publications

References 21 publications

Multi-institutional development and external validation of machine learning-based models to predict relapse risk of pancreatic ductal adenocarcinoma after radical resection

Multi-institutional development and external validation of machine learning-based models to predict relapse risk of pancreatic ductal adenocarcinoma after radical resection

Achieving ‘Marginal Gains’ to Optimise Outcomes in Resectable Pancreatic Cancer

Perioperative Predictors of Early Recurrence for Resectable and Borderline-Resectable Pancreatic Cancer

Contact Info

Product

Resources

About