A Machine-Based Approach to Preoperatively Identify Patients with the Most and Least Benefit Associated with Resection for Intrahepatic Cholangiocarcinoma: An International Multi-institutional Analysis of 1146 Patients

Impact of clinical lymph node status on survival in patients with intrahepatic cholangiocarcinoma undergoing liver resection plus lymphadenectomy

BACKGROUNDS

Liver resection plus lymphadenectomy is essential to ensure precise staging in patients with intrahepatic cholangiocarcinoma (ICC). This study aimed to investigate the influence of the clinical status of lymph nodes on the survival outcomes in ICC patients.

METHODS

Between January 2015 and December 2020, consecutive patients diagnosed with ICC who underwent liver resection plus lymphadenectomy were enrolled. Clinical assessment of lymph node status included positron emission tomography/computed tomography examination by radiologists pre-operatively, alongside intraoperative abdominal examination by the surgical team. Retrospective collection and analysis of clinical information alongside survival data were performed to assess outcomes.

RESULTS

The study included a total of 359 patients, with 291 (81.0%) and 151 (42.1%) displaying clinically and pathologically positive lymph nodes, respectively. The clinical assessment method had a sensitivity of 81.2% and a specificity of 54.3%. Following a median follow-up period of 32 months, the overall survival (OS) rates at 1, 3, and 5 years were 69.1%, 50.6%, and 41.2%, respectively, while the disease-free survival (DFS) rates were 60.7%, 42.8%, and 40.1%, respectively, across the cohort. Patients who had clinically positive but pathologically negative lymph nodes recorded the highest median OS (52 months) and median DFS (32 months). Conversely, those who were clinically negative but pathologically positive experienced the lowest median OS (16 months) and median DFS (8 months).

CONCLUSION

The current approach to clinically assessing lymph node status in ICC has a significant rate of false positives. Patients with clinically positive but pathologically negative lymph nodes exhibit the most favourable survival outcomes.

Application of AI on cholangiocarcinoma

Cholangiocarcinoma, classified as intrahepatic, perihilar, and extrahepatic, is considered a deadly malignancy of the hepatobiliary system. Most cases of cholangiocarcinoma are asymptomatic. Therefore, early detection of cholangiocarcinoma is significant but still challenging. The routine screening of a tumor lacks specificity and accuracy. With the application of AI, high-risk patients can be easily found by analyzing their clinical characteristics, serum biomarkers, and medical images. Moreover, AI can be used to predict the prognosis including recurrence risk and metastasis. Although they have some limitations, AI algorithms will still significantly improve many aspects of cholangiocarcinoma in the medical field with the development of computing power and technology.

Machine-Based Learning Hierarchical Cluster Analysis: Sex-Based Differences in Prognosis Following Resection of Hepatocellular Carcinoma

BACKGROUND

Patients with hepatocellular carcinoma (HCC) may have a heterogeneous presentation, as well as different long-term outcomes following surgical resection. We sought to use machine learning to cluster patients into different prognostic groups based on preoperative characteristics.

METHODS

Patients who underwent curative-intent liver resection for HCC between 2000 and 2020 were identified from a large international multi-institutional database. A hierarchical cluster analysis was performed based on preoperative factors to characterize patterns of presentation and define disease-free survival (DFS).

RESULTS

Among 966 with HCC, 3 distinct clusters were identified: Cluster 1 (n = 160, 16.5%), Cluster 2 (n = 537, 55.6%) and Cluster 3 (n = 269, 27.8%). Cluster 1 (n = 160, 16.5%) consisted of female patients (n = 160, 100%), low inflammation-based scores, intermediate tumor burden score (TBS) (median: 4.71) and high alpha-fetoprotein (AFP) levels (median 41.3 ng/mL); Cluster 2 consisted of male individuals (n = 537, 100%), mainly with a history of HBV infection (n = 429, 79.9%), low inflammation-based scores, intermediate AFP levels (median 26.0 ng/mL) and lower TBS (median 4.49); Cluster 3 was comprised of older patients (median age 68 years) predominantly male (n = 248, 92.2%) who had low incidence of HBV/HCV infection (7.1% and 8.2%, respectively), intermediate AFP levels (median 16.8 ng/mL), high inflammation-based scores and high TBS (median 6.58). Median DFS worsened incrementally among the three different clusters with Cluster 3 having the lowest DFS (Cluster 1: median not reached; Cluster 2: 34 months, 95% CI 23.0-48.0, Cluster 3: 19 months, 95% CI 15.0-29.0, p < 0.05).

CONCLUSION

Cluster analysis classified HCC patients into three distinct prognostic groups. Cluster assignment predicted DFS following resection of HCC with the female cluster having the most favorable prognosis following HCC resection.

Surgical Management of Intrahepatic Cholangiocarcinoma: Quo Vadis

Intrahepatic cholangiocarcinoma (ICC) is the second most common primary liver malignancy related to very high mortality rates [...].

Development and Validation of a Machine-Learning Model to Predict Early Recurrence of Intrahepatic Cholangiocarcinoma

BACKGROUND

The high incidence of early recurrence after hepatectomy for intrahepatic cholangiocarcinoma (ICC) has a detrimental effect on overall survival (OS). Machine-learning models may improve the accuracy of outcome prediction for malignancies.

METHODS

Patients who underwent curative-intent hepatectomy for ICC were identified using an international database. Three machine-learning models were trained to predict early recurrence (< 12 months after hepatectomy) using 14 clinicopathologic characteristics. The area under the receiver operating curve (AUC) was used to assess their discrimination ability.

RESULTS

In this study, 536 patients were randomly assigned to training (n = 376, 70.1%) and testing (n = 160, 29.9%) cohorts. Overall, 270 (50.4%) patients experienced early recurrence (training: n = 150 [50.3%] vs testing: n = 81 [50.6%]), with a median tumor burden score (TBS) of 5.6 (training: 5.8 [interquartile range {IQR}, 4.1-8.1] vs testing: 5.5 [IQR, 3.7-7.9]) and metastatic/undetermined nodes (N1/NX) in the majority of the patients (training: n = 282 [75.0%] vs testing n = 118 [73.8%]). Among the three different machine-learning algorithms, random forest (RF) demonstrated the highest discrimination in the training/testing cohorts (RF [AUC, 0.904/0.779] vs support vector machine [AUC, 0.671/0.746] vs logistic regression [AUC, 0.668/0.745]). The five most influential variables in the final model were TBS, perineural invasion, microvascular invasion, CA 19-9 lower than 200 U/mL, and N1/NX disease. The RF model successfully stratified OS relative to the risk of early recurrence.

CONCLUSIONS

Machine-learning prediction of early recurrence after ICC resection may inform tailored counseling, treatment, and recommendations. An easy-to-use calculator based on the RF model was developed and made available online.

Prediction of angiogenesis in extrahepatic cholangiocarcinoma using MRI-based machine learning

Purpose

Reliable noninvasive method to preoperative prediction of extrahepatic cholangiocarcinoma (eCCA) angiogenesis are needed. This study aims to develop and validate machine learning models based on magnetic resonance imaging (MRI) for predicting vascular endothelial growth factor (VEGF) expression and the microvessel density (MVD) of eCCA.

Materials and methods

In this retrospective study from August 2011 to May 2020, eCCA patients with pathological confirmation were selected. Features were extracted from T1-weighted, T2-weighted, and diffusion-weighted images using the MaZda software. After reliability testing and feature screening, retained features were used to establish classification models for predicting VEGF expression and regression models for predicting MVD. The performance of both models was evaluated respectively using area under the curve (AUC) and Adjusted R-Squared (Adjusted R²).

Results

The machine learning models were developed in 100 patients. A total of 900 features were extracted and 77 features with intraclass correlation coefficient (ICC) < 0.75 were eliminated. Among all the combinations of data preprocessing methods and classification algorithms, Z-score standardization + logistic regression exhibited excellent ability both in the training cohort (average AUC = 0.912) and the testing cohort (average AUC = 0.884). For regression model, Z-score standardization + stochastic gradient descent-based linear regression performed well in the training cohort (average Adjusted R^2 = 0.975), and was also better than the mean model in the test cohort (average Adjusted R^2 = 0.781).

Conclusion

Two machine learning models based on MRI can accurately predict VEGF expression and the MVD of eCCA respectively.

Artificial intelligence to investigate predictors and prognostic impact of time to surgery in colon cancer

BACKGROUND AND OBJECTIVES

The role of time to surgery (TTS) for long-term outcomes in colon cancer (CC) remains ill-defined. We sought to utilize artificial intelligence (AI) to characterize the drivers of TTS and its prognostic impact.

METHODS

The National Cancer Database was utilized to identify patients diagnosed with non-metastatic CC between 2004 and 2018. AI models were employed to rank the importance of several sociodemographic, facility, and tumor characteristics in determining TTS, and postoperative survival.

RESULTS

Among 518 983 patients, 137 902 (26.6%) received intraoperative diagnosis of CC (TTS = 0), while 381 081 (74.4%) underwent elective surgery (TTS > 0) with median TTS of 19.0 days (interquartile range [IQR]: 7.0-33.0). An AI model, identified tumor stage, receipt of adequate lymphadenectomy, histologic grade, lymphovascular invasion, and insurance status as the most important variables associated with TTS = 0. Conversely, the type and location of treating facility and receipt of adjuvant therapy were among the most important variables for TTS > 0. Notably, TTS was among the most important variables associated with survival, and TTS > 3 weeks was associated with an incremental increase in mortality risk.

CONCLUSIONS

The identification of factors associated with TTS can help stratify patients most likely to suffer poor outcomes due to prolonged TTS, as well as guide quality improvement initiatives related to timely surgical care.

Efficacy and safety comparison of neoadjuvant chemotherapy followed by surgery and upfront surgery for treating intrahepatic cholangiocarcinoma: a systematic review and meta-analysis

BACKGROUND AND AIMS

Currently, surgical resection is the most commonly performed and effective treatment for intrahepatic cholangiocarcinoma (ICC) worldwide. However, the prognosis of ICC is unsatisfactory. This study aimed to compare the efficacy and safety of neoadjuvant chemotherapy followed by surgery and upfront surgery in treating intrahepatic cholangiocarcinoma (ICC). The study also intends to explore whether chemotherapy should be introduced before surgery and which populations should be considered for neoadjuvant chemotherapy.

METHOD

Four databases, including PubMed, EMBASE, Cochrane Library, and Web of Science, were searched from their inception dates to January 2022 for relevant articles. The statistical analysis was performed using the Review Manager Software (version5.3). The non-randomized interventions (ROBINS-I) was used to assess the methodological quality of included studies and the overall quality of evidence was assessed through the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) tool. Moreover, the primary outcomes included 1-year, 3-year and 5-year overall survival (OS), while the secondary outcomes were R0 resection, 1-year, 3-year and 5-year recurrence-free survival (RFS), postoperative complications and ninety-day postoperative mortality.

RESULTS

Five studies involving 2412 patients were included in this meta-analysis. There was no significant difference in 1-year OS, 3-year OS, 1-year, 3-year and 5-year RFS, postoperative complications and ninety-day postoperative mortality between the two groups. However, the meta-analysis showed that the neoadjuvant chemotherapy group had a better 5-year OS benefit in ICC patients than the upfront surgery group (OR = 1.27, 95% CI: 1.02-1.58), while the R0 resection rate was lower in neoadjuvant chemotherapy group than that in the upfront surgery group (OR = 0.49, 95% CI: 0.26-0.91).

CONCLUSION

Compared with the upfront surgery, neoadjuvant chemotherapy followed by surgery could prolong the 5-year OS without increasing the risk of postoperative complications in ICC patients. Considering that the patients in the neoadjuvant chemotherapy followed by surgery group had more advanced ICC cases, the benefits of neoadjuvant chemotherapy may be more significant in patients with more advanced ICC.

Prognostic influence of multiple hepatic lesions in resectable intrahepatic cholangiocarcinoma: A systematic review and meta-analysis

BACKGROUND

Presence of multiple hepatic lesions in intrahepatic cholangiocarcinoma (iCCA) is included in staging as a negative prognostic factor, but both prognostic value and therapeutic implications remain debated. The aim of this study was to systematically review the prognostic influence of multiple lesions on survival after resection for iCCA, with stratification for distribution and number of lesions.

METHODS

Medline and Embase were systematically searched to identify records (2010-2021) reporting survival for patients undergoing primary resection for iCCA. Included were original articles reporting overall survival, with data on multiple lesions including tumour distribution (satellites/other multiple lesions) and/or number. For meta-analysis, the random effects model and inverse variance method were used. PRISMA 2020 guidelines were followed.

RESULTS

Thirty-one studies were included for review. For meta-analysis, nine studies reporting data on the prognostic influence of satellite lesions (2737 patients) and six studies reporting data on multiple lesions other than satellites (1589 patients) were included. Satellite lesions (hazard ratio 1.89, 95% confidence interval 1.67-2.13) and multiple lesions other than satellites (hazard ratio 2.41, 95% confidence interval 1.72-3.37) were significant negative prognostic factors. Data stratified for tumour number, while limited, indicated increased risk per additional lesion.

CONCLUSION

Satellite lesions, as well as multiple lesions other than satellites, was a negative prognostic factor in resectable iCCA. Considering the prognostic impact, both tumour distribution and number of lesions should be evaluated together with other risk factors to allow risk stratification for iCCA patients with multiple lesions, rather than precluding resection for the entire patient group.

Preoperative Serum Markers and Risk Classification in Intrahepatic Cholangiocarcinoma: A Multicenter Retrospective Study

Accurate risk stratification selects patients who are expected to benefit most from surgery. This retrospective study enrolled 225 Japanese patients with intrahepatic cholangiocellular carcinoma (ICC) who underwent hepatectomy between January 2009 and December 2020 and identified preoperative blood test biomarkers to formulate a classification system that predicted prognosis. The optimal cut-off values of blood test parameters were determined by ROC curve analysis, with Cox univariate and multivariate analyses identifying prognostic factors. Risk classifications were established using classification and regression tree (CART) analysis. CART analysis revealed decision trees for recurrence-free survival (RFS) and overall survival (OS) and created three risk classifications based on machine learning of preoperative serum markers. Five-year rates differed significantly (p < 0.001) between groups: 60.4% (low-risk), 22.8% (moderate-risk), and 4.1% (high-risk) for RFS and 69.2% (low-risk), 32.3% (moderate-risk), and 9.2% (high-risk) for OS. No difference in OS was observed between patients in the low-risk group with or without postoperative adjuvant chemotherapy, although OS improved in the moderate group and was prolonged significantly in the high-risk group receiving chemotherapy. Stratification of patients with ICC who underwent hepatectomy into three risk groups for RFS and OS identified preoperative prognostic factors that predicted prognosis and were easy to understand and apply clinically.

Prognostic value of lymphadenectomy in node-negative intrahepatic cholangiocarcinoma: A multicenter, retrospectively study

BACKGROUND

This study aimed to evaluate the prognostic value of lymph node dissection (LND) in node-negative intrahepatic cholangiocarcinoma (ICC) and identify the appropriately total number of lymph nodes examined (TNLE).

METHODS

Data from node-negative ICC patients who underwent curative intent resection in ten Chinese hepatobiliary centers from January 2010 to December 2018 were collected. Overall survival (OS), relapse-free survival (RFS) and postoperative complications were analyzed. Propensity score matching (PSM) was performed to reduce the bias due to confounding variables in LND group and non-lymph node dissection (NLND) group. The optimal TNLE was determined by survival analysis performed by the X-tile program using the enumeration method.

RESULTS

A total of 637 clinically node-negative ICC patients were included in this study, 74 cases were found lymph node (LN) positive after operation. Among the remaining 563 node-negative ICC patients, LND was associated with longer OS but not RFS before PSM (OS: 35.4 vs 26.0 months, p = 0.047; RFS: 15.0 vs 15.4 months, p = 0.992). After PSM, patients in LND group had better prognosis on both OS and RFS (OS: 38.0 vs 23.0 months, p < 0.001; RFS: 15.0 vs 13.0 months, p = 0.029). There were no statistically differences in postoperative complications. When TNLE was greater than 8, OS (48.5 vs 31.1 months, p = 0.025) and RFS (21.0 vs 13.0 months, p = 0.043) were longer in the group with more dissected LNs.

CONCLUSION

Routinely LND for node-negative ICC patients is recommended for it helps accurate tumor staging and associates with better prognosis. The optimal TNLE is more than 8.

Using machine learning to preoperatively stratify prognosis among patients with gallbladder cancer: a multi-institutional analysis

BACKGROUND

Gallbladder cancer (GBC) is an aggressive malignancy associated with a high risk of recurrence and mortality. We used a machine-based learning approach to stratify patients into distinct prognostic groups using preperative variables.

METHODS

Patients undergoing curative-intent resection of GBC were identified using a multi-institutional database. A classification and regression tree (CART) was used to stratify patients relative to overall survival (OS) based on preoperative clinical factors.

RESULTS

CART analysis identified tumor size, biliary drainage, carbohydrate antigen 19-9 (CA19-9) levels, and neutrophil-lymphocyte ratio (NLR) as the factors most strongly associated with OS. Machine learning cohorted patients into four prognostic groups: Group 1 (n = 109): NLR ≤1.5, CA19-9 ≤20, no drainage, tumor size <5.0 cm; Group 2 (n = 88): NLR >1.5, CA19-9 ≤20, no drainage, tumor size <5.0 cm; Group 3 (n = 46): CA19-9 >20, no drainage, tumor size <5.0 cm; Group 4 (n = 77): tumor size <5.0 cm with drainage OR tumor size ≥5.0 cm. Median OS decreased incrementally with CART group designation (59.5, 27.6, 20.6, and 12.1 months; p < 0.0001).

CONCLUSIONS

A machine-based model was able to stratify GBC patients into four distinct prognostic groups based only on preoperative characteristics. Characterizing patient prognosis with machine learning tools may help physicians provide more patient-centered care.

The prevalence and predictors of adjuvant chemotherapy use among patients treated with neoadjuvant endocrine therapy

PURPOSE

Neoadjuvant endocrine therapy (NET) facilitates clinical response and breast conservation in hormone receptor-positive (HR-positive) breast cancer. Patient selection for adjuvant chemotherapy (CT) post-NET is unclear and potentially evolving with use of genomic assays. We evaluated post-NET CT use in a national dataset.

METHODS

Using the National Cancer DataBase, we identified patients with cT2-3N0-3M0 HR-positive/human epidermal growth factor receptor 2-negative breast cancer treated between 2010 and 2017 with 3-12 months of NET prior to breast surgery. CT use was evaluated in the overall population, in patients with a pathologic complete response (pCR) and in patients with ypT1-2N0 disease (approximating PEPI 0). Exploratory analysis included patients > 50 years with ypN0-1, and 21-gene recurrence score (RS) ≤ 25 (approximating TAILORx/RxPONDER populations not benefiting from CT). Multivariable logistic regression was used to identify factors associated with CT.

RESULTS

Among 3624 eligible patients, 20.4% (740/3624) received CT. On multivariable analysis, age ≤ 50, lobular histology, grade 2, progesterone receptor negativity, ypT3, ypN + and RS ≥ 18 were associated with CT receipt. Co-morbidity, longer NET duration, ypT4, ypNx, and RS < 18 were associated with CT omission. CT was administered to 3.3% (1/30) of patients experiencing pCR and 5.5% (82/1483) with ypT1-2N0 disease. Among patients > 50 years with ypT0-3N0-1 residual disease, 13.8% (355/2569) received CT; RS was available for 24.8% (88/355) and 60% (53/88) had a score 0-25.

CONCLUSION

A minority of patients receive CT post-NET. This decision appears to be driven by younger age, RS and pathological nodal status. Increased consideration of these factors prior to neoadjuvant treatment choice may be warranted.

Advancements of Artificial Intelligence in Liver-Associated Diseases and Surgery

Background and Objectives: The advancement of artificial intelligence (AI) based technologies in medicine is progressing rapidly, but the majority of its real-world applications has not been implemented. The establishment of an accurate diagnosis with treatment has now transitioned into an artificial intelligence era, which has continued to provide an amplified understanding of liver cancer as a disease and helped to proceed better with the method of procurement. This article focuses on reviewing the AI in liver-associated diseases and surgical procedures, highlighting its development, use, and related counterparts. Materials and Methods: We searched for articles regarding AI in liver-related ailments and surgery, using the keywords (mentioned below) on PubMed, Google Scholar, Scopus, MEDLINE, and Cochrane Library. Choosing only the common studies suggested by these libraries, we segregated the matter based on disease. Finally, we compiled the essence of these articles under the various sub-headings. Results: After thorough review of articles, it was observed that there was a surge in the occurrence of liver-related surgeries, diagnoses, and treatments. Parallelly, advanced computer technologies governed by AI continue to prove their efficacy in the accurate screening, analysis, prediction, treatment, and recuperation of liver-related cases. Conclusions: The continual developments and high-order precision of AI is expanding its roots in all directions of applications. Despite being novel and lacking research, AI has shown its intrinsic worth for procedures in liver surgery while providing enhanced healing opportunities and personalized treatment for liver surgery patients.

Development and validation of a gradient boosting machine to predict prognosis after liver resection for intrahepatic cholangiocarcinoma

Background

Accurate prognosis assessment is essential for surgically resected intrahepatic cholangiocarcinoma (ICC) while published prognostic tools are limited by modest performance. We therefore aimed to establish a novel model to predict survival in resected ICC based on readily-available clinical parameters using machine learning technique.

Methods

A gradient boosting machine (GBM) was trained and validated to predict the likelihood of cancer-specific survival (CSS) on data from a Chinese hospital-based database using nested cross-validation, and then tested on the Surveillance, Epidemiology, and End Results (SEER) database. The performance of GBM model was compared with that of proposed prognostic score and staging system.

Results

A total of 1050 ICC patients (401 from China and 649 from SEER) treated with resection were included. Seven covariates were identified and entered into the GBM model: age, tumor size, tumor number, vascular invasion, number of regional lymph node metastasis, histological grade, and type of surgery. The GBM model predicted CSS with C-Statistics ≥ 0.72 and outperformed proposed prognostic score or system across study cohorts, even in sub-cohort with missing data. Calibration plots of predicted probabilities against observed survival rates indicated excellent concordance. Decision curve analysis demonstrated that the model had high clinical utility. The GBM model was able to stratify 5-year CSS ranging from over 54% in low-risk subset to 0% in high-risk subset.

Conclusions

We trained and validated a GBM model that allows a more accurate estimation of patient survival after resection compared with other prognostic indices. Such a model is readily integrated into a decision-support electronic health record system, and may improve therapeutic strategies for patients with resected ICC.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-022-09352-3.

Prediction Efficacy of Prognostic Nutritional Index and Albumin-Bilirubin Grade in Patients With Intrahepatic Cholangiocarcinoma After Radical Resection: A Multi-Institutional Analysis of 535 Patients

Background

The preoperative nutritional status and the immunological status have been reported to be independent prognostic factors of patients with intrahepatic cholangiocarcinoma (ICC). This study aimed to investigate whether prognostic nutritional index (PNI) + albumin–bilirubin (ALBI) could be a better predictor than PNI and ALBI alone in patients with ICC after radical resection.

Methods

The prognostic prediction evaluation of the PNI, ALBI, and the PNI+ALBI grade was performed in 373 patients with ICC who underwent radical resection between 2010 and 2018 at six Chinese tertiary hospitals, and external validation was conducted in 162 patients at four other Chinese tertiary hospitals. Overall survival (OS) and relapse-free survival (RFS) were estimated using the Kaplan–Meier method. Multivariate analysis was conducted to identify independent prognostic factors. A time-dependent receiver operating characteristic (ROC) curve and a nomogram prediction model were further constructed to assess the predictive ability of PNI, ALBI, and the PNI+ALBI grade. The C-index and a calibration plot were used to assess the performance of the nomogram models.

Results

Univariate analysis showed that PNI, ALBI, and the PNI+ALBI grade were prognostic factors for the OS and RFS of patients with ICC after radical resection in the training and testing sets (p < 0.001). Multivariate analysis showed that the PNI+ALBI grade was an independent risk factor for OS and RFS in the training and testing sets (p < 0.001). Analysis of the relationship between the PNI+ALBI grade and clinicopathological characteristics showed that the PNI+ALBI grade correlated with obstructive jaundice, alpha-fetoprotein (AFP), cancer antigen 19-9 (CA19-9), cancer antigen 125 (CA125), PNI, ALBI, Child–Pugh grade, type of resection, tumor size, major vascular invasion, microvascular invasion, T stage, and N stage (p < 0.05). The time-dependent ROC curves showed that the PNI+ALBI grade had better prognostic predictive ability than the PNI, ALBI, and the Child–Pugh grade in the training and testing sets.

Conclusion

Preoperative PNI+ALBI grade is an effective and practical predictor for the OS and RFS of patients with ICC after radical resection.

Machine learning to guide clinical decision-making in abdominal surgery-a systematic literature review

PURPOSE

An indication for surgical therapy includes balancing benefits against risk, which remains a key task in all surgical disciplines. Decisions are oftentimes based on clinical experience while guidelines lack evidence-based background. Various medical fields capitalized the application of machine learning (ML), and preliminary research suggests promising implications in surgeons' workflow. Hence, we evaluated ML's contemporary and possible future role in clinical decision-making (CDM) focusing on abdominal surgery.

METHODS

Using the PICO framework, relevant keywords and research questions were identified. Following the PRISMA guidelines, a systemic search strategy in the PubMed database was conducted. Results were filtered by distinct criteria and selected articles were manually full text reviewed.

RESULTS

Literature review revealed 4,396 articles, of which 47 matched the search criteria. The mean number of patients included was 55,843. A total of eight distinct ML techniques were evaluated whereas AUROC was applied by most authors for comparing ML predictions vs. conventional CDM routines. Most authors (N = 30/47, 63.8%) stated ML's superiority in the prediction of benefits and risks of surgery. The identification of highly relevant parameters to be integrated into algorithms allowing a more precise prognosis was emphasized as the main advantage of ML in CDM.

CONCLUSIONS

A potential value of ML for surgical decision-making was demonstrated in several scientific articles. However, the low number of publications with only few collaborative studies between surgeons and computer scientists underpins the early phase of this highly promising field. Interdisciplinary research initiatives combining existing clinical datasets and emerging techniques of data processing may likely improve CDM in abdominal surgery in the future.

Intrahepatic Cholangiocarcinoma: A Summative Review of Biomarkers and Targeted Therapies

Simple Summary

Intrahepatic cholangiocarcinoma is the second most common primary liver malignancy. Among patients with operable disease, surgical resection is the cornerstone of therapy. Among the majority of patients who present with advanced disease treatment, systemic or targeted therapy is indicated. Recent advancements have provided more novel therapeutic approaches to a subset of patients with intrahepatic cholangiocarcinoma.

Abstract

Although rare, intrahepatic cholangiocarcinoma (ICC) is the second most common primary hepatic malignancy and the incidence of ICC has increased 14% per year in recent decades. Treatment of ICC remains difficult as most people present with advanced disease not amenable to curative-intent surgical resection. Even among patients with operable disease, margin-negative surgical resection can be difficult to achieve and the incidence of recurrence remains high. As such, there has been considerable interest in systemic chemotherapy and targeted therapy for ICC. Over the last decade, the understanding of the molecular and genetic foundations of ICC has reshaped treatment approaches and strategies. Next-generation sequencing has revealed that most ICC tumors have at least one targetable mutation. These advancements have led to multiple clinical trials to examine the safety and efficacy of novel therapeutics that target tumor-specific molecular and genetic aberrations. While these advancements have demonstrated survival benefit in early phase clinical trials, continued investigation in randomized larger-scale trials is needed to further define the potential clinical impact of such therapy.

Challenges and opportunities in the application of artificial intelligence in gastroenterology and hepatology

Artificial intelligence (AI) is an umbrella term used to describe a cluster of interrelated fields. Machine learning (ML) refers to a model that learns from past data to predict future data. Medicine and particularly gastroenterology and hepatology, are data-rich fields with extensive data repositories, and therefore fruitful ground for AI/ML-based software applications. In this study, we comprehensively review the current applications of AI/ML-based models in these fields and the opportunities that arise from their application. Specifically, we refer to the applications of AI/ML-based models in prevention, diagnosis, management, and prognosis of gastrointestinal bleeding, inflammatory bowel diseases, gastrointestinal premalignant and malignant lesions, other nonmalignant gastrointestinal lesions and diseases, hepatitis B and C infection, chronic liver diseases, hepatocellular carcinoma, cholangiocarcinoma, and primary sclerosing cholangitis. At the same time, we identify the major challenges that restrain the widespread use of these models in healthcare in an effort to explore ways to overcome them. Notably, we elaborate on the concerns regarding intrinsic biases, data protection, cybersecurity, intellectual property, liability, ethical challenges, and transparency. Even at a slower pace than anticipated, AI is infiltrating the healthcare industry. AI in healthcare will become a reality, and every physician will have to engage with it by necessity.

Computer auxiliary diagnosis technique of detecting cholangiocarcinoma based on medical imaging: A review

BACKGROUND AND OBJECTIVES

Cholangiocarcinoma (CCA) is one of the most aggressive human malignant tumors and is becoming one of the main factors of death and disability globally. Specifically, 60% to 70% of CCA patients were diagnosed with local invasion or distant metastasis and lost the chance of radical operation. The overall median survival time was less than 12 months. As a non-invasive diagnostic technology, medical imaging consisting of computed tomography (CT) imaging, magnetic resonance imaging (MRI), and ultrasound (US) imaging, is the most effectively and commonly used method to detect CCA. The computer auxiliary diagnosis (CAD) system based on medical imaging is helpful for rapid diagnosis and provides credible "second opinion" for specialists. The purpose of this review is to categorize and review the CAD technique of detecting CCA based on medical imaging.

METHODS

This work applies a four-level screening process to choose suitable publications. 125 research papers published in different academic research databases were selected and analyzed according to specific criteria. From the five steps of medical image acquisition, processing, analysis, understanding and verification of CAD combined with artificial intelligence algorithms, we obtain the most advanced insights related to CCA detection.

RESULTS

This work provides a comprehensive analysis and comparison analysis of the current CAD systems of detecting CCA. After careful investigation, we find that the main detection methods are traditional machine learning method and deep learning method. For the detection, the most commonly used method is semi-automatic segmentation algorithm combined with support vector machine classifier method, combination of which has good detection performance. The end-to-end training mode makes deep learning method more and more popular in CAD systems. However, due to the limited medical training data, the accuracy of deep learning method is unsatisfactory.

CONCLUSIONS

Based on analysis of artificial intelligence methods applied in CCA, this work is expected to be truly applied in clinical practice in the future to improve the level of clinical diagnosis and treatment of it. This work concludes by providing a prediction of future trends, which will be of great significance for researchers in the medical imaging of CCA and artificial intelligence.

Reappraisal of the T Category for Solitary Intrahepatic Cholangiocarcinoma by Tumor Size in 611 Early-Stage (T1-2N0M0) Patients After Hepatectomy: a Surveillance, Epidemiology, and End Results (SEER) Analysis

BACKGROUND

The association between tumor size and survival in patients with intrahepatic cholangiocarcinoma (ICC) after hepatectomy is controversial, and the T category in the American Joint Committee on Cancer (AJCC) stage for ICC is a topic of debate.

METHODS

Data from 611 T1-2N0M0 ICC patients classified by the AJCC 8th edition who underwent hepatectomy were extracted from the Surveillance, Epidemiology, and End Results (SEER) database during 1988-2015. Cancer-specific survival was evaluated using Kaplan-Meier analysis. The optimal cutoff value of solitary tumor size was used an adjusted p value approach to discriminating patient survival.

RESULTS

In the AJCC 8th staging system, using a 5-cm cut-off value of tumor size for solitary ICC without vascular invasion (S/VI-) was not associated with survival in T1 category (p = 0.201), and multifocal ICC with vascular invasion had a worse survival than solitary ICC with vascular invasion (S/VI+) in T2 category (p = 0.014). Tumor size was a prognostic factor for both S/VI- and S/VI+, the optimal cutoff value of tumor size was obtained 8 cm for S/VI- and 3 cm for S/VI+. S/VI- ≤ 8 cm had a similar survival to S/VI+ ≤ 3 cm (p = 0.126), S/VI- > 8 cm had a similar survival to S/VI+ > 3 cm (p = 0.655), and multifocal ICC had a similar survival with S/VI- > 8 cm (p = 0.159) and S/VI+ > 3 cm (p = 0.196). When the cohort was divided into two groups-new T1 (S/VI- ≤ 8 cm and S/VI+ ≤ 3 cm) and new T2 (S/VI- > 8 cm, S/VI+ > 3 cm and multifocal ICC)-significant survival difference was observed (p < 0.0001).

CONCLUSIONS

The discriminatory power of the AJCC 8th edition for solitary ICC could be further enhanced by subdividing tumors according to size and vascular invasion (8 cm for S/VI- and 3 cm for S/VI+).

Predicting Lymph Node Metastasis in Intrahepatic Cholangiocarcinoma

BACKGROUND

The objective of the current study was to develop a model to predict the likelihood of occult lymph node metastasis (LNM) prior to resection of intrahepatic cholangiocarcinoma (ICC).

METHODS

Patients who underwent hepatectomy for ICC between 2000 and 2017 were identified using a multi-institutional database. A novel model incorporating clinical and preoperative imaging data was developed to predict LNM.

RESULTS

Among 980 patients who underwent resection of ICC, 190 (19.4%) individuals had at least one LNM identified on final pathology. An enhanced imaging model incorporating clinical and imaging data was developed to predict LNM ( https://k-sahara.shinyapps.io/ICC_imaging/ ). The performance of the enhanced imaging model was very good in the training data set (c-index 0.702), as well as the validation data set with bootstrapping resamples (c-index 0.701) and outperformed the preoperative imaging alone (c-index 0.660). The novel model predicted both 5-year overall survival (OS) (low risk 48.4% vs. high risk 18.4%) and 5-year disease-specific survival (DSS) (low risk 51.9% vs. high risk 25.2%, both p < 0.001). When applied among Nx patients, 5-year OS and DSS of low-risk Nx patients was comparable with that of N0 patients, while high-risk Nx patients had similar outcomes to N1 patients (p > 0.05).

CONCLUSION

This tool may represent an opportunity to stratify prognosis of Nx patients and can help inform clinical decision-making prior to resection of ICC.

Therapeutic Outcomes and Prognostic Factors of Unresectable Intrahepatic Cholangiocarcinoma: A Data Mining Analysis

Prognosis of patients with intrahepatic cholangiocarcinoma (ICC) is unsatisfactory. Tumor, host, and treatment factors including hepatic arterial infusion chemotherapy (HAIC) are intricately involved in the progression of ICC. We aimed to identify profiles associated with disease control rate (DCR) and the prognosis of patients with unresectable ICC by decision tree analysis. We analyzed 31 consecutive patients with unresectable ICC (median age, 71 years; the male ratio was 58.1%). Stage IVB occupied 51.6% of patients, and 38.7% and 58.1% of patients were treated with gemcitabine plus cisplatin combination therapy and HAIC, respectively. Profiles associated with prognosis as well as DCR were investigated by decision tree analysis. The median survival time (MST) of the patients was 11.6 months, and the DCR was 70.9%. Multivariate correlation analysis showed that albumin levels and WBC levels were significantly correlated with survival time (albumin, ρ = 0.3572, p = 0.0485; WBC, ρ = -0.4008, p = 0.0280). In decision tree analysis, WBC level was selected as the initial split variable, and subjects with WBC levels of 6800/μL or less (45.1%) showed a long survival time (MST 476 days). We also demonstrated that the profile associated with the highest DCR was "less than 4.46 mg/dL of CRP levels and treatment with HAIC". We demonstrated a new prognostic profile for ICC patients, which consisted of WBC and CRP levels. Moreover, we demonstrated that HAIC was associated with better disease control in ICC patients with low CPR levels. Thus, these new profiles may be useful for the management of ICC patients.

Very Early Recurrence After Liver Resection for Intrahepatic Cholangiocarcinoma: Considering Alternative Treatment Approaches

Importance

Although surgery offers the best chance of a potential cure for patients with localized, resectable intrahepatic cholangiocarcinoma (ICC), prognosis of patients remains dismal largely because of a high incidence of recurrence.

Objective

To predict very early recurrence (VER) (ie, recurrence within 6 months after surgery) following resection for ICC in the pre- and postoperative setting.

Design, Setting, and Participants

Patients who underwent curative-intent resection for ICC between May 1990 and July 2016 were identified from an international multi-institutional database. The study was conducted at The Ohio State University in collaboration with all other participating institutions. The data were analyzed in December 2019.

Main Outcomes and Measures

Two logistic regression models were constructed to predict VER based on pre- and postoperative variables. The final models were used to develop an online calculator to predict VER and the tool was internally and externally validated.

Results

Among 880 patients (median age, 59 years [interquartile range, 51-68 years]; 388 women [44.1%]; 428 [50.2%] white; 377 [44.3%] Asian; 27 [3.2%] black]), 196 (22.3%) developed VER. The 5-year overall survival among patients with and without VER was 8.9% vs 49.8%, respectively (P < .001). A preoperative model was able to stratify patients relative to the risk for VER: low risk (6-month recurrence-free survival [RFS], 87.7%), intermediate risk (6-month RFS, 72.3%), and high risk (6-month RFS, 49.5%) (log-rank P < .001). The postoperative model similarly identified discrete cohorts of patients based on probability for VER: low risk (6-month RFS, 90.0%), intermediate risk (6-month RFS, 73.1%), and high risk (6-month RFS, 48.5%) (log-rank, P < .001). The calibration and predictive accuracy of the pre- and postoperative models were good in the training (C index: preoperative, 0.710; postoperative, 0.722) as well as the internal (C index: preoperative, 0.715; postoperative, 0.728; bootstrapping resamples, n = 5000) and external (C index: postoperative, 0.672) validation data sets.

Conclusion and Relevance

An easy-to-use online calculator was developed to help clinicians predict the chance of VER after curative-intent resection for ICC. The tool performed well on internal and external validation. This tool may help clinicians in the preoperative selection of patients for neoadjuvant therapy as well as during the postoperative period to inform surveillance strategies.

Artificial intelligence in transplantation (machine-learning classifiers and transplant oncology)

PURPOSE OF REVIEW

To highlight recent efforts in the development and implementation of machine learning in transplant oncology - a field that uses liver transplantation for the treatment of hepatobiliary malignancies - and particularly in hepatocellular carcinoma, the most commonly treated diagnosis in transplant oncology.

RECENT FINDINGS

The development of machine learning has occurred within three domains related to hepatocellular carcinoma: identification of key clinicopathological variables, genomics, and image processing.

SUMMARY

Machine-learning classifiers can be effectively applied for more accurate clinical prediction and handling of data, such as genetics and imaging in transplant oncology. This has allowed for the identification of factors that most significantly influence recurrence and survival in disease, such as hepatocellular carcinoma, and thus help in prognosticating patients who may benefit from a liver transplant. Although progress has been made in using these methods to analyse clinicopathological information, genomic profiles, and image processed data (both histopathological and radiomic), future progress relies on integrating data across these domains.

Tumor Burden Dictates Prognosis Among Patients Undergoing Resection of Intrahepatic Cholangiocarcinoma: A Tool to Guide Post-Resection Adjuvant Chemotherapy?

INTRODUCTION

While tumor burden (TB) has been associated with outcomes among patients with hepatocellular carcinoma, the role of overall TB in intrahepatic cholangiocarcinoma (ICC) remains poorly defined.

METHODS

Patients undergoing curative-intent resection of ICC between 2000 and 2017 were identified from a multi-institutional database. The impact of TB on overall (OS) and disease-free survival (DFS) was evaluated in the multi-institutional database and validated externally.

RESULTS

Among 1101 patients who underwent curative-intent resection of ICC, 624 (56.7%) had low TB, 346 (31.4%) medium TB, and 131 (11.9%) high TB. OS incrementally worsened with higher TB (5-year OS; low TB: 48.3% vs medium TB: 29.8% vs high TB: 17.3%, p < 0.001). Similarly, patients with low TB had better DFS compared with medium and high TB patients (5-year DFS: 38.3% vs 18.7% vs 6.9%, p < 0.001). On multivariable analysis, TB was independently associated with OS (medium TB: HR = 1.40, 95% CI 1.14-1.71; high TB: HR = 1.89, 95% CI 1.46-2.45) and DFS (medium TB, HR = 1.61, 95% CI 1.33-1.96; high TB: HR = 2.03, 95% CI 1.56-2.64). Survival analysis revealed an excellent prognostic discrimination using the TB among the external validation cohort (3-year OS; low TB: 44.8%, medium TB: 29.3%; high TB: 23.3%, p = 0.03; 3-year DFS: low TB: 32.7%, medium TB: 10.7%; high TB: 0%, p < 0.001). While neoadjuvant chemotherapy was not associated with survival across the TB groups, receipt of adjuvant chemotherapy was associated with increased survival among patients with high TB (5-year OS: 24.4% vs 13.4%, p = 0.02).

CONCLUSION

Overall TB dictated prognosis among patients with resectable ICC. TB may be used as a tool to help guide post-resection treatment strategies.

The systemic immune-inflammation index predicts prognosis in intrahepatic cholangiocarcinoma: an international multi-institutional analysis

BACKGROUND

The objective of this study was to examine whether the systemic immune inflammation index (SII) was associated with prognosis among patients following resection of intrahepatic cholangiocarcinoma (ICC).

METHODS

The impact of SII on overall (OS) and cancer-specific survival (CSS) following resection of ICC was assessed. The performance of the final multivariable models that incorporated inflammatory markers (i.e. neutrophil-to-lymphocyte ratio [NLR], platelet-to-lymphocyte ratio [PLR] and SII [platelets∗NLR]) was assessed using the Harrell's concordance index.

RESULTS

Patients with high SII had worse 5-year OS (37.7% vs 46.6%, p < 0.001) and CSS (46.1% vs 50.1%, p < 0.001) compared with patients with low SII. An elevated SII (HR = 1.70, 95% CI 1.23-2.34) and NLR (HR = 1.58, 95% CI 1.10-2.27) independently predicted worse OS, whereas high PLR (HR = 1.17, 95% CI 0.85-1.60) was no longer associated with prognosis. Only SII remained an independent predictor of CSS (HR = 1.55, 95% CI 1.09-2.21). The SII multivariable model outperformed models that incorporated PLR and NLR relative to OS (c-index; 0.696 vs 0.689 vs 0.692) and CSS (c-index; 0.697 vs 0.689 vs 0.690).

CONCLUSION

SII independently predicted OS and CSS among patients with resectable ICC. SII may be a better predictor of outcomes compared with other markers of inflammatory response among patients with resectable ICC.

Prognostic role of artificial intelligence among patients with hepatocellular cancer: A systematic review

BACKGROUND

Prediction of survival after the treatment of hepatocellular carcinoma (HCC) has been widely investigated, yet remains inadequate. The application of artificial intelligence (AI) is emerging as a valid adjunct to traditional statistics due to the ability to process vast amounts of data and find hidden interconnections between variables. AI and deep learning are increasingly employed in several topics of liver cancer research, including diagnosis, pathology, and prognosis.

AIM

To assess the role of AI in the prediction of survival following HCC treatment.

METHODS

A web-based literature search was performed according to the Preferred Reporting Items for Systemic Reviews and Meta-Analysis guidelines using the keywords “artificial intelligence”, “deep learning” and “hepatocellular carcinoma” (and synonyms). The specific research question was formulated following the patient (patients with HCC), intervention (evaluation of HCC treatment using AI), comparison (evaluation without using AI), and outcome (patient death and/or tumor recurrence) structure. English language articles were retrieved, screened, and reviewed by the authors. The quality of the papers was assessed using the Risk of Bias In Non-randomized Studies of Interventions tool. Data were extracted and collected in a database.

RESULTS

Among the 598 articles screened, nine papers met the inclusion criteria, six of which had low-risk rates of bias. Eight articles were published in the last decade; all came from eastern countries. Patient sample size was extremely heterogenous (n = 11-22926). AI methodologies employed included artificial neural networks (ANN) in six studies, as well as support vector machine, artificial plant optimization, and peritumoral radiomics in the remaining three studies. All the studies testing the role of ANN compared the performance of ANN with traditional statistics. Training cohorts were used to train the neural networks that were then applied to validation cohorts. In all cases, the AI models demonstrated superior predictive performance compared with traditional statistics with significantly improved areas under the curve.

CONCLUSION

AI applied to survival prediction after HCC treatment provided enhanced accuracy compared with conventional linear systems of analysis. Improved transferability and reproducibility will facilitate the widespread use of AI methodologies.

Neoadjuvant treatment strategies for intrahepatic cholangiocarcinoma

Intrahepatic cholangiocarcinoma (ICC) is the second most common primary liver malignancy and is increasing in incidence. Long-term outcomes are optimized when patients undergo margin-negative resection followed by adjuvant chemotherapy. Unfortunately, a significant proportion of patients present with locally advanced, unresectable disease. Furthermore, recurrence rates are high even among patients who undergo surgical resection. The delivery of systemic and/or liver-directed therapies prior to surgery may increase the proportion of patients who are eligible for surgery and reduce recurrence rates by prioritizing early systemic therapy for this aggressive cancer. Nevertheless, the available evidence for neoadjuvant therapy in ICC is currently limited yet recent advances in liver directed therapies, chemotherapy regimens, and targeted therapies have generated increasing interest its role. In this article, we review the rationale for, current evidence for, and ongoing research efforts in the use of neoadjuvant therapy for ICC.

Part 1: Artificial intelligence technology in surgery

Artificial intelligence (AI) is one of the disruptive technologies of the fourth Industrial Revolution that is changing our work practices. This technology is in use in highly diverse industries including health care, defence, insurance and e-commerce. This review focuses on the relevance of AI to surgery. AI will aid surgeons with diagnostic decision-making, patient selection for surgery as well as improve patient pre- and post-operative care and management. Ethical considerations of AI with respect to patient rights and data privacy are highlighted. A further challenge is how best to present to national regulators a pragmatic way to assess AI as 'software as a medical device'. This relates to the ramifications for the adoption of AI technology in clinical practice, and its subsequent public funding support and reimbursement. It is evident that AI technology has important applications in surgery in the 21st century. The establishment of a key work programme in this area will be important if surgeons are to fully utilize AI in surgery.

Preoperative leukocytosis and the resection severity index are independent risk factors for survival in patients with intrahepatic cholangiocarcinoma

PURPOSE

The incidence of intrahepatic cholangiocarcinoma is increasing worldwide. Despite advances in surgical and non-surgical treatment, reported outcomes are still poor and surgical resection remains to be the only chance for long-term survival of affected patients. The identification and validation of prognostic factors and scores, such as the recently introduced resection severity index, for postoperative morbidity and mortality are essential to facilitate optimal therapeutic regimens.

METHODS

This is a retrospective analysis of 269 patients undergoing resection of histologically confirmed intrahepatic cholangiocarcinoma between February 1996 and September 2018 at a tertiary referral center for hepatobiliary surgery. Regression analyses were performed to evaluate potential prognostic factors, including the resection severity index.

RESULTS

Median postoperative follow-up time was 22.93 (0.10-234.39) months. Severe postoperative complications (≥ Clavien-Dindo grade III) were observed in 94 (34.9%) patients. The body mass index (p = 0.035), the resection severity index (ASAT in U/l divided by Quick in % multiplied by the extent of liver resection graded in points; p = 0.006), additional hilar bile duct resection (p = 0.005), and number of packed red blood cells transfused during operation (p = 0.036) were independent risk factors for the onset of severe postoperative complications. Median Kaplan-Meier survival after resection was 27.63 months. Preoperative leukocytosis (p = 0.003), the resection severity index (p = 0.005), multivisceral resection (p = 0.001), and T stage ≥ 3 (p = 0.013) were identified as independent risk factors for survival.

CONCLUSION

Preoperative leukocytosis and the resection severity index are useful variables for preoperative risk stratification since they were identified as significant predictors for postoperative morbidity and mortality, respectively.

Cholangiocarcinoma: investigations into pathway-targeted therapies

INTRODUCTION

Cholangiocarcinoma is a malignant disease of the biliary tract and accounts for 3% of all gastrointestinal tumors. Surgical intervention is currently the only potentially curative strategy for cholangiocarcinoma. For patients with unresectable, advanced or metastatic disease, the combination of gemcitabine with cisplatin is considered the standard treatment. However, currently available therapeutic options have only a marginal benefit, especially among patients with relapsed/refractory tumors.

AREAS COVERED

We reviewed targeted agents under clinical evaluation for patients with cholangiocarcinoma. FGFR and IDH inhibitors are at the most advanced stage of clinical investigation. EGFR inhibitors have demonstrated contradictory results, whereas inhibition of other molecular pathways, including the RAS/RAF/MEK/ERK, the MET, the PI3K/AKT/mTOR and angiogenetic pathways, has shown minimal or null benefit.

EXPERT OPINION

Several targeted approaches are being investigated for advanced cholangiocarcinoma. However, randomized clinical trials are needed to define the optimal treatment regimen and address issues including the option of monotherapy or combination regimens, the optimal sequence of different treatments, ways to overcome resistance to targeted treatments, as well as determining the right time and tissue for assessing molecular signatures. Targeted therapies and immunotherapy hold promise for improving patient outcomes in the future.

A Novel Classification of Intrahepatic Cholangiocarcinoma Phenotypes Using Machine Learning Techniques: An International Multi-Institutional Analysis

INTRODUCTION

Patients with intrahepatic cholangiocarcinoma (ICC) generally have a poor prognosis, yet there can be heterogeneity in the patterns of presentation and associated outcomes. We sought to identify clusters of ICC patients based on preoperative characteristics that may have distinct outcomes based on differing patterns of presentation.

METHODS

Patients undergoing curative-intent resection of ICC between 2000 and 2017 were identified using a multi-institutional database. A cluster analysis was performed based on preoperative variables to identify distinct patterns of presentation. A classification tree was built to prospectively assign patients into cluster assignments.

RESULTS

Among 826 patients with ICC, three distinct presentation patterns were noted. Specifically, Cluster 1 (common ICC, 58.9%) consisted of individuals who had a small-size ICC (median 4.6 cm) and median carbohydrate antigen (CA) 19-9 and neutrophil-to-lymphocyte ratio (NLR) levels of 40.3 UI/mL and 2.6, respectively; Cluster 2 (proliferative ICC, 34.9%) consisted of patients who had larger-size tumors (median 9.0 cm), higher CA19-9 levels (median 72.0 UI/mL), and similar NLR (median 2.7); Cluster 3 (inflammatory ICC, 6.2%) comprised of patients with a medium-size ICC (median 6.2 cm), the lowest range of CA19-9 (median 26.2 UI/mL), yet the highest NLR (median 13.5) (all p < 0.05). Median OS worsened incrementally among the three different clusters {Cluster 1 vs. 2 vs. 3; 60.4 months (95% confidence interval [CI] 43.0-77.8) vs. 27.2 months (95% CI 19.9-34.4) vs. 13.3 months (95% CI 7.2-19.3); p < 0.001}. The classification tree used to assign patients into different clusters had an excellent agreement with actual cluster assignment (κ = 0.93, 95% CI 0.90-0.96).

CONCLUSION

Machine learning analysis identified three distinct prognostic clusters based solely on preoperative characteristics among patients with ICC. Characterizing preoperative patient heterogeneity with machine learning tools can help physicians with preoperative selection and risk stratification of patients with ICC.