Predictive Nomogram for Recurrence of Stage I Colorectal Cancer After Curative Resection

Machine learning for predicting colon cancer recurrence

INTRODUCTION

Colorectal cancer (CRC) is a global public health concern, ranking among the most commonly diagnosed malignancies worldwide. Despite advancements in treatment modalities, the specter of CRC recurrence remains a significant challenge, demanding innovative solutions for early detection and intervention. The integration of machine learning into oncology offers a promising avenue to address this issue, providing data-driven insights and personalized care.

METHODS

This retrospective study analyzed data from 396 patients who underwent surgical procedures for colon cancer (CC) between 2010 and 2021. Machine learning algorithms were employed to predict CC recurrence, with a focus on demographic, clinicopathological, and laboratory characteristics. A range of evaluation metrics, including AUC (Area Under the Receiver Operating Characteristic), accuracy, recall, precision, and F1 scores, assessed the performance of machine learning algorithms.

RESULTS

Significant risk factors for CC recurrence were identified, including sex, carcinoembryonic antigen (CEA) levels, tumor location, depth, lymphatic and venous invasion, and lymph node involvement. The CatBoost Classifier demonstrated exceptional performance, achieving an AUC of 0.92 and an accuracy of 88 % on the test dataset. Feature importance analysis highlighted the significance of CEA levels, albumin levels, N stage, weight, platelet count, height, neutrophil count, lymphocyte count, and gender in determining recurrence risk.

DISCUSSION

The integration of machine learning into healthcare, exemplified by this study's findings, offers a pathway to personalized patient risk stratification and enhanced clinical decision-making. Early identification of individuals at risk of CC recurrence holds the potential for more effective therapeutic interventions and improved patient outcomes.

CONCLUSION

Machine learning has the potential to revolutionize our approach to CC recurrence prediction, emphasizing the synergy between medical expertise and cutting-edge technology in the fight against cancer. This study represents a vital step toward precision medicine in CC management, showcasing the transformative power of data-driven insights in oncology.

Determinants of survival and recurrence in patients with stage I colorectal cancer

Background

Due to the novel advanced screening methods, the number of patients diagnosed with stage I colorectal cancer (CRC) is increasing. This retrospective cohort study aimed to identify recurrence and survival risk factors of patients with stage I CRC after surgery.

Materials and methods

Patients with stage I CRC were evaluated, and their demographic and clinicopathologic variables were recorded. The log-rank test assessed the association of variables with overall survival (OS), recurrence-free survival (RFS), local recurrence, and distant metastasis.

Results

The median overall survival period was 51 months. The recurrence rate was 13.7%: 7.2% local and 9.3% distant recurrence. One-, two-, three-, and five-year RFS were 92%, 89%, 87%, and 83%, respectively, and OS were 96%, 93%, 90%, and 89%, respectively. Local and distant recurrence rates were higher in patients with higher tumor grades. Additionally, RFS and OS were worse in patients with higher tumor grades, and perforation was associated with worse OS.

Conclusions

The determinants of survival and recurrence identified in the present study can be used to improve patient outcomes by early diagnosis and appropriate management of high-risk patients.

Analysis of risk factors for postoperative recurrence of stage I colorectal cancer: a retrospective analysis of a large population

Background

Colorectal cancer (CRC) is the third most common cancer worldwide. Patients diagnosed with stage I CRC typically do not require postoperative adjuvant treatment. However, postoperative recurrence is present in at least 40% of patients with CRC and often occurs in those with stage I disease. This study aimed to elucidate the current status of recurrence and clinicopathological characteristics in patients with stage I CRC.

Methods

Data of indicated patients were obtained from 18 registries in Surveillance, Epidemiology, and End Results (SEER). The multivariable Fine-Gray regression model was used to identify the mortality risk of patients. Disparities in survival were analyzed using Kaplan-Meier curves. Logistic regression was employed to identify factors associated with recurrent risk overestimation.

Results

Our study indicated a recurrence rate of 15.04% (1,874/12,452) in stage I CRC cases. Notably, we identified race, age, T stage, and carcinoembryonic antigen (CEA) levels as independent risk factors for tumor recurrence, substantially impacting prognosis. Furthermore, gender, race (Black), age (>65 years), elevated CEA levels, and refusal or unknown status regarding radiotherapy significantly correlated with an adverse prognosis in patients with stage I CRC.

Conclusions

We identified certain key clinicopathological features of patients with stage I CRC and demonstrated the survival benefits of radiotherapy, offering a new perspective on stage I CRC follow-up and treatment recommendations.

KCTD15 acts as an anti-tumor factor in colorectal cancer cells downstream of the demethylase FTO and the m6A reader YTHDF2

Potassium Channel Tetramerization Domain Containing 15 (KCTD15) participates in the carcinogenesis of several solid malignancies; however, its role in colorectal cancer (CRC) remains unclear. Here we find that KCTD15 exhibits lower expression in CRC tissues as compared to para-carcinoma tissues. Tetracycline (tet)-induced overexpression and knockdown of KCTD15 confirms KCTD15 as an anti-proliferative and pro-apoptotic factor in CRC both in vitro and in xenografted tumors. N6-methyladenosine (m6A) is known to affect the expression, stabilization, and degradation of RNAs with this modification. We demonstrate that upregulation of fat mass and obesity-associated protein (FTO), a classical m6A eraser, prevents KCTD15 mRNA degradation in CRC cells. Less KCTD15 RNA is recognized by m6A 'reader' YTH N6-Methyladenosine RNA Binding Protein F2 (YTHDF2) in FTO-overexpressed cells. Moreover, KCTD15 overexpression decreases protein expression of histone deacetylase 1 (HDAC1) but increases acetylation of critical tumor suppressor p53 at Lys373 and Lys382. Degradation of p53 is delayed in CRC cells post-KCTD15 overexpression. We further show that the regulatory effects of KCTD15 on p53 are HDAC1-dependent. Collectively, we conclude that KCTD15 functions as an anti-growth factor in CRC cells, and its expression is orchestrated by the FTO-YTHDF2 axis. Enhanced p53 protein stabilization may contribute to KCTD15's actions in CRC cells.

Prediction of disease recurrence or residual disease after primary endoscopic resection of pT1 colorectal cancer-results from a large nationwide Danish study

PURPOSE

Risk assessment of disease recurrence in pT1 colorectal cancer is crucial in order to select the appropriate treatment strategy. The study aimed to develop a prediction model, based on histopathological data, for the probability of disease recurrence and residual disease in patients with pT1 colorectal cancer.

METHODS

The model dataset consisted of 558 patients with pT1 CRC who had undergone endoscopic resection only (n = 339) or endoscopic resection followed by subsequent bowel resection (n = 219). Tissue blocks and slides were retrieved from Pathology Departments from all regions in Denmark. All original slides were evaluated by one experienced gastrointestinal pathologist (TPK). New sections were cut and stained for haematoxylin and eosin (HE) and immunohistochemical markers. Missing values were multiple imputed. A logistic regression model with backward elimination was used to construct the prediction model.

RESULTS

The final prediction model for disease recurrence demonstrated good performance with AUC of 0.75 [95% CI 0.72-0.78], HL chi-squared test of 0.59 and scaled Brier score of 10%. The final prediction model for residual disease demonstrated medium performance with an AUC of 0.68 [0.63-0.72].

CONCLUSION

We developed a prediction model for the probability of disease recurrence in pT1 CRC with good performance and calibration based on histopathological data. Together with lymphatic and venous invasion, an involved resection margin (0 mm) as opposed to a margin of ≤ 1 mm was an independent risk factor for both disease recurrence and residual disease.

Prognostic impact of tumor size on patients with metastatic colorectal cancer: a large SEER-based retrospective cohort study

Given the poor prognosis of metastatic colorectal cancer (mCRC), this research aimed to investigate the correlation between tumor size and prognosis, and develop a novel prediction model to guide individualized treatment. Patients pathologically diagnosed with mCRC were recruited from the Surveillance, Epidemiology, and End Results (SEER) database between 2010 and 2015, and were randomly divided (7:3 ratio) into a training cohort (n = 5597) and a validation cohort (n = 2398). Kaplan-Meier curves were used to analyze the relationship between tumor size and overall survival (OS). Univariate Cox analysis was applied to assess the factors associated with the prognosis of mCRC patients in the training cohort, and then multivariate Cox analysis was used to construct a nomogram model. The area under the receiver-operating characteristics curve (AUC) and calibration curve were used to evaluate the predictive ability of the model. Patients with larger tumors had a worse prognosis. While brain metastases were associated with larger tumors compared to liver or lung metastases, bone metastases tended to be associated with smaller tumors. Multivariate Cox analysis revealed that tumor size was an independent prognostic risk factor (HR 1.28, 95% CI 1.19-1.38), in addition to the other ten variables (age, race, primary site, grade, histology, T stage, N stage, chemotherapy, CEA level and metastases site). The 1-, 3-, and 5-year OS nomogram model yielded AUC values of more than 0.70 in both the training and validation cohorts, and its predictive performance was superior to that of the traditional TNM stage. Calibration plots demonstrated a good agreement between the predicted and observed 1-, 3-, and 5-year OS outcomes in both cohorts. The size of primary tumor was found to be significantly associated with prognosis of mCRC, and was also correlated with specific metastatic organ. In this study, we presented the first effort to create and validate a novel nomogram for predicting 1-, 3- and 5-year OS probabilities of mCRC. The prognostic nomogram was demonstrated to have an excellent predictive ability in estimating individualized OS of patients with mCRC.

Risk factors of recurrence in TNM stage I colorectal cancer

Purpose

TNM stage I colorectal cancer (CRC) can recur, although the recurrence rate is low. Few studies have evaluated the risk factors for TNM stage I CRC recurrence. This study aimed to evaluate the TNM stage I CRC recurrence rate, as well as risk factors for recurrence.

Methods

In this retrospective study, we reviewed the database of patients who had undergone surgery for TNM stage I CRC between November 2008 and December 2014 without receiving neoadjuvant therapy or transanal excision for rectal cancer. Our analysis included 173 patients. Primary lesions were found in the colon of 133 patients and in the rectum of 40 patients.

Results

The CRC recurrence rate was 2.9% (5 out of 173 patients). For colon cancer patients, tumor size was not associated with higher recurrence risk (P = 0.098). However, for rectal cancer patients, both tumor size (≥3 cm) and T stage were associated with higher recurrence risk (P = 0.046 and P = 0.046, respectively). Of the 5 recurrent cases, 1 patient exhibited disease progression despite treatment, 1 patient maintained stable disease status after recurrence treatment, and 3 patients had no evidence of a tumor after recurrence treatment.

Conclusion

Our findings suggest that tumor size and T stage are predictors of stage I rectal cancer recurrence, and careful monitoring and follow-up of patients with larger tumors may be warranted.

Nomogram for prediction of prolonged postoperative ileus after colorectal resection

BACKGROUND

Prolonged postoperative ileus (PPOI) is a major complication in patients undergoing colorectal resection. The aim of this study was to analyze the risk factors contributing to PPOI, and to develop an effective nomogram to determine the risks of this population.

METHODS

A total of 1,254 patients with colorectal cancer who underwent radical colorectal resection at Fujian Cancer Hospital from March 2016 to August 2021 were enrolled as a training cohort in this study. Univariate analysis and multivariate logistic regressions were performed to determine the correlation between PPOI and clinicopathological characteristics. A nomogram predicting the incidence of PPOI was constructed. The cohort of 153 patients from Fujian Provincial Hospital were enrolled as a validation cohort. Internal and external validations were used to evaluate the prediction ability by area under the receiver operating characteristic curve (AUC) and a calibration plot.

RESULTS

In the training cohort, 128 patients (10.2%) had PPOI after colorectal resection. The independent predictive factors of PPOI were identified, and included gender, age, surgical approach and intraoperative fluid overload. The AUC of nomogram were 0.779 (95% CI: 0.736-0.822) and 0.791 (95%CI: 0.677-0.905) in the training and validation cohort, respectively. The two cohorts of calibration plots showed a good consistency between nomogram prediction and actual observation.

CONCLUSIONS

A highly accurate nomogram was developed and validated in this study, which can be used to provide individual prediction of PPOI in patients after colorectal resection, and this predictive power can potentially assist surgeons to make the optimal treatment decisions.

Effects of deficient mismatch repair on the prognosis of patients with stage II and stage III colon cancer during different postoperative periods

Background

We evaluated the prognostic role of deficient mismatch repair (dMMR) systems in stage II and stage III colon cancer patients during different postoperative periods. We also assessed whether patients aged ≥75 could benefit from chemotherapy.

Methods

This retrospective study was conducted across three medical centers in China. Kaplan–Meier survival methods and Cox proportional hazards models were used to evaluate the differences in overall survival (OS) and disease-free survival (DFS) rates. Propensity score matching was performed to reduce imbalances in the baseline characteristics of the patients. Landmark analysis was performed to evaluate the role of dMMR during different postoperative periods.

Results

The median follow-up time for all patients was 45.0 months (25–75 IQR: 38.0–82.5). There was no significant OS (p = 0.350) or DFS (p = 0.752) benefit associated with dMMR for stage II and III patients during the first postoperative year. However, significant OS (p < 0.001) and DFS (p < 0.001) benefits were observed from the second postoperative year until the end of follow-up. These differences remained after propensity score matching. Moreover, chemotherapy produced no OS (HR = 0.761, 95% CI: 0.43–1.34, p = 0.341) or DFS (HR = 0.98, 95% CI: 0.51–1.88, p = 0.961) benefit for patients aged ≥75 years.

Conclusion

The benefits of dMMR in stage III patients were observed from the second postoperative year until the end of follow-up. However, the prognosis of patients with dMMR is not different from that of patients with proficient mismatch repair (pMMR) during the first postoperative year. In addition, elderly patients aged ≥75 years obtained no significant survival benefits from postoperative chemotherapy.

Prediction of Metachronous Peritoneal Metastases After Radical Surgery for Colon Cancer: A Scoring System Obtained from an International Multicenter Cohort

Background

Since novel strategies for prevention and treatment of metachronous peritoneal metastases (mPM) are under study, it appears crucial to identify their risk factors. Our aim is to establish the incidence of mPM after surgery for colon cancer (CC) and to build a statistical model to predict the risk of recurrence.

Patients and Methods

Retrospective analysis of consecutive pT3–4 CC operated at five referral centers (2014–2018). Patients who developed mPM were compared with patients who were PM-free at follow-up. A scoring system was built on the basis of a logistic regression model.

Results

Of the 1423 included patients, 74 (5.2%) developed mPM. Patients in the PM group presented higher preoperative carcinoembryonic antigen (CEA) [median (IQR): 4.5 (2.5–13.0) vs. 2.7 (1.5–5.9), P = 0.001] and CA 19-9 [median (IQR): 17.7 (12.0–37.0) vs. 10.8 (5.0–21.0), P = 0.001], advanced disease (pT4a 42.6% vs. 13.5%; pT4b 16.2% vs. 3.2%; P < 0.001), and negative pathological characteristics. Multivariate logistic regression identified CA 19-9, pT stage, pN stage, extent of lymphadenectomy, and lymphovascular invasion as significant predictors, and individual risk scores were calculated for each patient. The risk of recurrence increased remarkably with score values, and the model demonstrated a high negative predictive value (98.8%) and accuracy (83.9%) for scores below five.

Conclusions

Besides confirming incidence and risk factors for mPM, our study developed a useful clinical tool for prediction of mPM risk. After external validation, this scoring system may guide personalized decision-making for patients with locally advanced CC.

Supplementary Information

The online version contains supplementary material available at 10.1245/s10434-022-12097-9.

LINC00022 acts as an oncogene in colorectal cancer progression via sponging miR-375-3p to regulate FOXF1 expression

Background

Abnormal expression of long non-coding RNAs (lncRNAs) has been shown to be associated with the pathogenesis of cancers, including colorectal cancer (CRC). It has been reported that LINC00022 is highly expressed in some typs of cancer and its overexpression indicates poor prognosis. The function of LINC00022 in CRC progression remains unclear and is mainly investigated in the present study.

Methods

LINC00022 expression in CRC tissues was analyzed by using the TNMplot software. LINC00022 expression in CRC cells was measured by quantitative real-time PCR. The effects of LINC00022 on the malignant behaviors of CRC cells were detected by a series of in vitro and in vivo experiments. Dual-luciferase assays were used to verify the targeting relationship between LINC00022 and miR-375-3p and between miR-375-3p and Forkhead box F1 (FOXF1), followed by the rescue experiment.

Results

LINC00022 was highly expressed in CRC tissues compared with paired para-carcinoma tissues (n = 41). CRC cells with LINC00022 knockdown exhibited decreased cell proliferation, migration, and invasion abilities but increased apoptosis accompanied by decreased protein levels of c-Myc, cyclin D1, cleaved caspase 3, cleaved poly(ADP-ribose) polymerase, matrix metalloproteinase (MMP) 2, and MMP9. Additionally, LINC00022 downregulation in CRC cells suppressed the tube formation of human umbilical vein endothelial cells (HUVECs) as evidenced by decreased vascular endothelial growth factor A levels in LINC00022-silenced cells. The inhibitory effect of LINC00022 knockdown on tumor growth was also observed in an in vivo model. Conversely, LINC00022 overexpression showed that opposite effect. We further demonsrtaed that LINC00022 could upregulate FOXF1 expression through sponging miR-375-3p. Moreover, miR-375-3p knockdown reversed the effects of LINC00022 down-regulation.

Conclusions

LINC00022 may up-regulate FOXF1 expression via competitively binding miR-375-3p, thereby promoting the development of CRC.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-022-09566-5.

Predicting metachronous liver metastasis in patients with colorectal cancer: development and assessment of a new nomogram

Background

We aimed to develop and validate a nomogram model, which could predict metachronous liver metastasis in colorectal cancer within two years after diagnosis.

Methods

A retrospective study was performed on colorectal cancer patients who were admitted to Beijing Shijitan Hospital from January 1, 2016 to June 30, 2019. The least absolute shrinkage and selection operator (LASSO) regression model was used to optimize feature selection for susceptibility to metachronous liver metastasis in colorectal cancer. Multivariable logistic regression analysis was applied to establish a predictive model through incorporating features selected in the LASSO regression model. C-index, receiver operating characteristic (ROC) curve, calibration plot, and decision curve analysis (DCA) were employed to assess discrimination, distinctiveness, consistency with actual occurrence risk, and clinical utility of candidate predictive model. Internal validation was assessed with bootstrapping method.

Results

Predictors contained in candidate prediction nomogram included age, CEA, vascular invasion, T stage, N stage, family history of cancer, and KRAS mutation. This model displayed good discrimination with a C-index of 0.787 (95% confidence interval: 0.728–0.846) and good calibration, whereas area under the ROC curve (AUC) of 0.786. Internal validation obtained C-index of 0.786, and AUC of validation cohort is 0.784. Based on DCA, with threshold probability range from 1 to 60%; this predictive model might identify colorectal cancer metachronous liver metastasis to achieve a net clinical benefit.

Conclusion

We have developed and validated a prognostic nomogram with good discriminative and high accuracy to predict metachronous liver metastasis in CRC patients.

Predictive and Prognostic Effects of Primary Tumor Size on Colorectal Cancer Survival

Background

Pathologic staging is crucial in colorectal cancer (CRC). Unlike the majority of solid tumors, the current staging model does not use tumor size as a criterion. We evaluated the predictive and prognostic impact of primary tumor size on all stages of CRC.

Methods

Using the National Cancer Database (NCDB), we conducted an analysis of CRC patients diagnosed between 2010 and 2015 who underwent resection of their primary cancer. Univariate and multivariate analyses were used to identify predictive and prognostic factors, Kaplan-Meier analysis and Cox proportional hazards models for association between tumor size and survival.

Results

About 61,000 patients met the inclusion criteria. Median age was 63 years and majority of the tumors were colon primary (82.7%). AJCC stage distribution was: I - 20.1%; II - 32.1%; III - 34.7% and IV - 13.1%. The prognostic impact of tumor size was strongly associated with survival in stage III disease. Compared to patients with tumors <2cm; those with 2-5cm (HR 1.33; 1.19-1.49; p<0.001), 5-10cm (HR 1.51 (1.34-1.70; p<0.001) and >10cm (HR 1.95 (1.65-2.31; p<0.001) had worse survival independent of other variables. Stage II treated without adjuvant chemotherapy had comparable survival outcomes (HR 1.09; 0.97-1.523; p=0.148) with stage III patients who did, while Stage II patients who received adjuvant chemotherapy did much better than both groups (HR 0.76; 0.67-0.86; p<0.001). Stage III patients who did not receive adjuvant chemotherapy had the worst outcomes among the non-metastatic disease subgroups (HR 2.66; 2.48-2.86; p<0.001). Larger tumors were associated with advanced stage, MSI high, non-rectal primary and positive resection margins.

Conclusions

Further studies are needed to clarify the role of tumor size in prognostic staging models, and how to incorporate it into therapy decisions.

Perineural invasion and number of retrieved lymph nodes are prognostic factors for T2N0 colon cancer

PURPOSE

The prognosis of pathological T2N0 colon cancer has not been adequately investigated. This study aimed to determine the prognostic factors for pathological T2N0 colon cancer by comparing it with those for pathological T3N0 colon cancer.

METHODS

We retrospectively reviewed patients with primary colon cancer who underwent curative resection between January 2007 and December 2015 and included 889 patients with postoperative pathological T2-3N0M0 disease. The clinicopathological characteristics were analyzed to identify the independent prognostic factors.

RESULTS

Pathological T2 (n = 185, 20.8%) and T3 (n = 704, 79.2%) tumors showed no difference in the 5-year disease-free survival (5Y DFS) rate (95.8% vs. 93.2%, p = 0.257) after a median follow-up of 55 months (range, 1-106 months). Multivariate Cox regression analysis showed that perineural invasion (hazard ratio [HR] = 2.041, 95% confidence interval [CI] 1.122-3.712, p = 0.019) and number of retrieved lymph nodes < 12 (HR = 2.994, 95% CI 1.327-6.753, p = 0.008) were independent prognostic factors for DFS. Pathological T2 tumors with poor prognostic factors showed similar 5Y DFS as that of T3 tumors with poor prognostic factors (88.9% vs. 88.6%, p = 0.916), but not with T3 tumors without poor prognostic factors (88.9% vs. 95.0%, p = 0.089).

CONCLUSION

Pathological T2N0 colon cancer showed oncologic outcomes similar to that of T3N0 colon cancer. Therefore, more intensive surveillance is necessary for patients with high-risk T2N0 colon cancer.

The Spectrum, Tendency and Predictive Value of PIK3CA Mutation in Chinese Colorectal Cancer Patients

Background

PIK3CA is a high-frequency mutation gene in colorectal cancer, while its prognostic value remains unclear. This study evaluated the mutation tendency, spectrum, prognosis power and predictive power in cetuximab treatment of PIK3CA in Chinese CRC cohort.

Methods

The PIK3CA exon 9 and 20 status of 5763 CRC patients was detected with Sanger sequencing and a high-resolution melting test. Clinicopathological characteristics of 5733 patients were analyzed. Kaplan-Meier method and nomogram were used to evaluate the overall survival curve and disease recurrence, respectively.

Results

Fifty-eight types of mutations in 13.4% (771/5733) of the patients were detected. From 2014 to 2018, the mutation rate of PIK3CA increased from 11.0% to 13.5%. At stage IV, exon 20 mutated patients suffered shorter overall survival time than wild-type patients (multivariate COX regression analysis, HR = 2.72, 95% CIs = 1.47-5.09; p-value = 0.012). At stage III, PIK3CA mutated patients were more likely to relapse (multivariate Logistic regression analysis, exon 9: OR = 2.54, 95% CI = 1.34-4.73, p = 0.003; exon 20: OR = 3.89, 95% CI = 1.66-9.10, p = 0.002). The concordance index of the nomogram for predicting the recurrence risk of stage III patients was 0.685. After cetuximab treatment, the median PFS of PIK3CA exon 9 wild-type patients (n = 9) and mutant patients (n = 5) did not reach a significant difference (3.6 months vs. 2.3 months, Log-rank test, p-value = 0.513).

Conclusions

We found that PIK3CA mutation was an adverse predictive marker for the overall survival of stage IV patients and recurrence of stage III patients, respectively. Further more, we suggested that PIK3CA exon 9 mutations are not negative predictors of cetuximab treatment in KRAS, NRAS, and BRAF wild-type mCRC patients.

Disease recurrence after colorectal cancer surgery in the modern era: a population-based study

PURPOSE

This population-based study determined the cumulative incidence (CI) of local, regional, and distant recurrences, examined metastatic patterns, and identified risk factors for recurrence after curative treatment for CRC.

METHODS

All patients undergoing resection for pathological stage I-III CRC between January 2015 and July 2015 and registered in the Netherlands Cancer Registry were selected (N = 5412). Additional patient record review and data collection on recurrences was conducted by trained administrators in 2019. Three-year CI of recurrence was calculated according to sublocation (right-sided: RCC, left-sided: LCC and rectal cancer: RC) and stage. Cox competing risk regression analyses were used to identify risk factors for recurrence.

RESULTS

The 3-year CI of recurrence for stage I, II, and III RCC and LCC was 0.03 vs. 0.03, 0.12 vs. 0.16, and 0.31 vs. 0.24, respectively. The 3-year CI of recurrence for stage I, II, and III RC was 0.08, 0.24, and 0.38. Distant metastases were found in 14, 12, and 16% of patients with RCC, LCC, and RC. Multiple site metastases were found often in patients with RCC, LCC, and RC (42 vs. 32 vs. 28%). Risk factors for recurrence in stage I-II CRC were age 65-74 years, pT4 tumor size, and poor tumor differentiation whereas in stage III CRC, these were ASA III, pT4 tumor size, N2, and poor tumor differentiation.

CONCLUSIONS

Recurrence rates in recently treated patients with CRC were lower than reported in the literature and the metastatic pattern and recurrence risks varied between anatomical sublocations.

Proposal for a post-operative surveillance strategy for stage I colorectal cancer patients based on a novel recurrence risk stratification: a multicenter retrospective study

BACKGROUND

The recurrent risk of stage I colorectal cancer (CRC) is not clear, and the data regarding appropriate post-operative surveillance schedules in stage I CRC are scarce.

OBJECTIVES

We aimed to stratify stage I CRC based on the recurrence risk and evaluate optimal post-operative surveillance durations based on this stratification.

METHODS

We retrospectively analyzed the cases of 6607 stage I CRC patients from 24 institutions. To assess the patients' clinicopathological factors that impact recurrence-free survival (RFS), we performed univariate and multivariate analyses using Cox proportional hazards models. We divided the patients into classes based on their numbers of factors that were associated with poor RFI in the multivariate analysis.

RESULTS

Recurrence occurred in 3.9% patients. The multivariate analysis revealed the independent factors for poor RFS: rectal cancer, T2 depth, presence of lymphatic invasion, high level of pre-operative carcinoembryonic antigen, and absence of D2-3 lymphadenectomy. We also divided the patients into three classes based on their numbers of these risk factors; the 3-year and 5-year RFS rates were 99.3% and 99.1% in the no-risk patients, 97.4% and 96.5% in the patients with 1-2 risks, and 92.1% and 90.0% in the patients with 3-5 risks, respectively. In the patients with no risk and in the patients with 1-2 risks after 3 years post-surgery, ≤ 1% recurrence occurred. Thus, post-operative surveillance may be omitted in these populations.

CONCLUSIONS

Our new classification properly stratified the recurrence risks of stage I CRC patients, and may help reduce unnecessary post-operative surveillance.

Development and validation of a Super learner-based model for predicting survival in Chinese Han patients with resected colorectal cancer

OBJECTIVE

Improved prognostic prediction for patients with colorectal cancer stays an important challenge. This study aimed to develop an effective prognostic model for predicting survival in resected colorectal cancer patients through the implementation of the Super learner.

METHODS

A total of 2333 patients who met the inclusion criteria were enrolled in the cohort. We used multivariate Cox regression analysis to identify significant prognostic factors and Super learner to construct prognostic models. Prediction models were internally validated by 10-fold cross-validation and externally validated with a dataset from The Cancer Genome Atlas. Discrimination and calibration were evaluated by Harrell concordence index (C-index) and calibration plots, respectively.

RESULTS

Age, T stage, N stage, histological type, tumor location, lymph-vascular invasion, preoperative carcinoembryonic antigen and sample lymph nodes were integrated into prediction models. The concordance index of Super learner-based prediction model (SLM) was 0.792 (95% confidence interval: 0.767-0.818), which is higher than that of the seventh edition American Joint Committee on Cancer TNM staging system 0.689 (95% confidence interval: 0.672-0.703) for predicting overall survival (P < 0.05). In the external validation, the concordance index of the SLM for predicting overall survival was also higher than that of tumor-node-metastasis (TNM) stage system (0.764 vs. 0.682, respectively; P < 0.001). In addition, the SLM showed good calibration properties.

CONCLUSIONS

We developed and externally validated an effective prognosis prediction model based on Super learner, which offered more reliable and accurate prognosis prediction and may be used to more accurately identify high-risk patients who need more active surveillance in patients with resected colorectal cancer.

Personalizing prognostic prediction in early-onset Colorectal Cancer

Accurately estimating prognosis based on clinicopathologic variables could improve risk stratification for patients with early-onset colorectal cancer (EOCRC). Our primary goal was to create and validate a survival nomogram with adequate performance for predicting overall survival (OS) in patients with EOCRC. Least absolute shrinkage and selection operator (LASSO) Cox regression analysis was applied to identify clinical features statistically related to OS. Then we established and internally validated a survival nomogram based on surveillance, epidemiology and end results (SEER) database (N=23813). A cohort of 77 patients with EOCRC from Renmin Hospital of Wuhan University (RHWU) was employed to detect the external validity of the survival nomogram. Moreover, we compared the predictive accuracy of survival nomogram with TNM stage, and also compared the OS between endoscopy and surgery groups before and after propensity score matching (PSM) among EOCRC patients with early stage (Tis-T1N0M0). We selected seven informative indexes (N stage, M stage, perineural invasion, chemotherapy, surgery primary site, summary stage and tumor grade) for the construction of the survival nomogram. Then the survival nomogram exhibited good discrimination with C-index of 0.829, 0.841 and 0.796 in the SEER training, SEER validation and RHWU validation sets, respectively. Calibration curves showed good concordance between the survival nomogram predictions and actual outcomes for 1-year, 3-year and 5-year OS. Furthermore, the survival nomogram was superior to risk stratification by TNM stage in predicting OS among patients with EOCRC. Early-stage patients treated with endoscopy showed similar survival to those with surgery before and after PSM. We proposed a survival nomogram based on the extensively used parameters to precisely predict OS in EOCRC patients. This survival nomogram will contribute to aid oncologists better risk stratification and prognostication for patients with EOCRC.

Risk factors and risk prediction models for colorectal cancer metastasis and recurrence: an umbrella review of systematic reviews and meta-analyses of observational studies

BACKGROUND

There is a clear need for systematic appraisal of models/factors predicting colorectal cancer (CRC) metastasis and recurrence because clinical decisions about adjuvant treatment are taken on the basis of such variables.

METHODS

We conducted an umbrella review of all systematic reviews of observational studies (with/without meta-analysis) that evaluated risk factors of CRC metastasis and recurrence. We also generated an updated synthesis of risk prediction models for CRC metastasis and recurrence. We cross-assessed individual risk factors and risk prediction models.

RESULTS

Thirty-four risk factors for CRC metastasis and 17 for recurrence were investigated. Twelve of 34 and 4/17 risk factors with p < 0.05 were estimated to change the odds of the outcome at least 3-fold. Only one risk factor (vascular invasion for lymph node metastasis [LNM] in pT1 CRC) presented convincing evidence. We identified 24 CRC risk prediction models. Across 12 metastasis models, six out of 27 unique predictors were assessed in the umbrella review and four of them changed the odds of the outcome at least 3-fold. Across 12 recurrence models, five out of 25 unique predictors were assessed in the umbrella review and only one changed the odds of the outcome at least 3-fold.

CONCLUSIONS

This study provides an in-depth evaluation and cross-assessment of 51 risk factors and 24 prediction models. Our findings suggest that a minority of influential risk factors are employed in prediction models, which indicates the need for a more rigorous and systematic model construction process following evidence-based methods.

A novel CpG-methylation-based nomogram predicts survival in colorectal cancer

Aberrant DNA methylation is significantly associated with the prognosis of patients with colorectal cancer (CRC). Therefore, the aim of this study was to develop a CpG-methylation-based nomogram for prognostic prediction in CRC. First, 378 CRC patients with methylation data from The Cancer Genome Atlas were randomly divided into training cohort (n = 249) and test cohort (n = 129). A multistep screening strategy was performed to identify six CpG sites that were significantly associated with overall survival in the training cohort. Then, Cox regression modelling was performed to construct a prognostic signature based on the candidate CpG sites. The six-CpG signature successfully separated patients into high-risk and low-risk groups in both training and test cohorts, and its performance was superior to that of previously published methylation markers (P < 0.05). Furthermore, we established a prognostic nomogram incorporating this signature, TNM stage, and age. The nomogram exhibited better prediction for overall survival in comparison with the three independent prognostic factors in the training cohort (C-index: 0.798 vs 0.620 to 0.737; P < 0.001). In the test cohort, the performance of nomogram was also superior to that of the three independent prognostic factors (C-index: 0.715 vs 0.590 to 0.665; P < 0.05). Meanwhile, the calibration curves for survival probability showed good agreement between prediction by nomogram and actual observation in both training and test cohorts. Together, the present study provides a novel CpG-methylation-based nomogram as a promising predictor for overall survival of CRC patients, which may help improve decision-making regarding the personalized treatments of patients with CRC.

STING signaling is a potential immunotherapeutic target in colorectal cancer

Background: Stimulator of Interferon Genes (STING) is an innate immune sensor for cytosolic DNA. STING signaling activation is indispensable for type I interferon response and the anti-cancer immune response by CD8⁺ T cells. The aim of this study was to characterize intratumoral STING expression pattern and its clinical implication in colorectal cancer (CRC). Methods: We analyzed STING and CD8 expression in 225 CRC patients who underwent surgical resection. Clinicopathological variables and survival outcomes were analyzed according to STING expression levels. Mice with syngeneic MC38 tumors were also treated with a STING agonist, and tumor microenvironments were analyzed using immunofluorescent staining and flow cytometry. Results: Distinct STING expression was observed in the CRC tumor specimens. Patients with higher STING expression had early stage cancer with increased intratumoral CD8⁺ T cell infiltration and less frequent lymphovascular invasion. Compared to CRC patients with lower STING expression, those with higher STING expression had longer overall and recurrence-free survival. Multivariate Cox regression model also revealed higher STING expression to be an independent prognostic factor for better overall survival. When MC38 colon tumors were treated with intratumoral injection of STING agonist, tumor growth was remarkably suppressed with increased intratumoral CD8⁺ T cell infiltration. Moreover, T-cell activation markers, ICOS and IFN-γ, were also upregulated in CD8⁺ T cells, indicating enhanced effector T cell function after STING treatment. Conclusion: We confirmed the distinct STING expression in CRC and demonstrated its independent prognostic value in survival outcomes. STING could be a potential therapeutic target that enhances anti-cancer immune response in CRC.