Predicting early extrahepatic recurrence after local treatment of colorectal liver metastases

Resectability assessment of colorectal liver metastases by an expert panel: Potential impact on hospitals referring patients for local treatment

BACKGROUND

Patients with colorectal liver-only metastases (CRLM) eligible for local treatment (resection/ablation) do not always receive this potentially curative treatment due to the lack of clear resectability criteria and expertise in centres not performing liver surgery. We evaluated the potential value of a liver expert panel in daily practice.

METHODS

All patients with CRLM starting with systemic treatment in centres not performing liver surgery between 2016 and 2020 were identified in the Netherlands Cancer Registry. A panel of liver surgeons retrospectively re-evaluated patients' imaging for resectability before and two-monthly during systemic treatment.

RESULTS

Sixty-three patients were included from 24 hospitals requiring a total of 544 resectability assessments by individual panel surgeons. The panel considered 18 (29 %) patients to have resectable CRLM before starting systemic treatment, which increased to 43 (68 %) after up to three evaluations. Eighteen (29 %) patients considered resectable by the panel at any time received no local treatment of whom 9 (50 %) were not referred to a liver surgeon.

CONCLUSION

In non-liver-surgery centres, over a quarter of patients technically eligible for local treatment of initially unresectable CRLM, sometimes mistakenly categorised as such, did not receive this. This stresses the need for liver expert panels in daily practice to increase local treatment rates.

The impact of neoadjuvant therapy on the prognostic value of preoperative neutrophil-to-lymphocyte ratio for colorectal liver metastases: a multi-center cohort study

BACKGROUND

Neutrophil-to-lymphocyte ratio (NLR) is a promising prognostic marker for patients undergoing hepatectomy for colorectal liver metastases (CRLM). However, its prognostic value in patients receiving neoadjuvant therapy (NAT) has not been sufficiently addressed.

METHODS

From 2013 to 2023, a cohort of 692 patients with CRLM receiving hepatectomy were enrolled in five centers. Clinicopathological characteristics were obtained from a prospectively maintained multi-center database. The effect of NLR (> 2 versus ≤ 2) on overall survival (OS) and recurrence-free survival was estimated by Kaplan-Meier analysis. Univariable and multivariable Cox regression analysis was applied to investigate the influence of individual clinicopathological parameters on OS.

RESULTS

In the entire cohort, the median NLR level was 2.11 (0.30-16.33). There were 307 (44.4%) patients receiving NAT followed by hepatectomy, while 385 (55.6%) patients undergoing upfront surgery. Notably, patients in the NAT group showed significantly lower NLR level than those in the upfront surgery group (1.83 versus 2.32, P < 0.001). In the upfront surgery group, high NLR was significantly associated with worse OS, independent of other factors (HR = 1.49, 95% CI 1.08-2.05, P = 0.02). In the NAT group, there was no significant difference in OS between the high NLR and low NLR group.

CONCLUSION

The prognostic value of NLR in surgically resected CRLM is potentially influenced by NAT in the modern era.

Patterns of recurrence after curative intent hepatic resection for colorectal liver metastasis

BACKGROUND

Improved surgical techniques and more intensive systemic therapy have increased the number of patients with oligometastatic colorectal cancer eligible for resection, but a significant percentage of these cases will ultimately recur. Furthermore, distinct recurrence patterns have been associated with different outcomes.

METHODS

Data for 195 patients who underwent curative-intent colorectal liver metastasis (CRLM) resection between 2016 and 2022 at Johns Hopkins Hospital were retrospectively collected. Cox regression univariate and multivariate analyses identified features associated with survival outcomes. Association between risk factors and site of recurrences was conducted via logistic regression with initial recurrences grouped into intrahepatic-only, extrahepatic-only, and concurrent intra- and extrahepatic recurrence.

RESULTS

The 1- and 2-year recurrence-free survival (RFS) rates were 46% and 22%, respectively. The 1- and 2-year overall survival (OS) rates were 95% and 88%, respectively. The median OS was 71.7 months. Multivariate analysis identified age <60 years, N2 nodal status, R1 liver margin, and higher preoperative carcinoembryonic antigen as top prognostic factors for worse RFS. Additionally, patients with left-sided primary tumors had a higher risk of intrahepatic-only recurrence, whereas mutant KRAS was associated with a higher risk of extrahepatic recurrence with or without liver recurrence.

CONCLUSION

These results from a recent cohort of patients treated with current standard-of-care therapies identify features associated with elevated risk and specific patterns of recurrence. These insights into CRLM recurrence patterns support a larger prospective study to identify subgroups of patients who may require additional therapies to prevent recurrence.

Prognostic value of radiologic and pathological response in colorectal cancer liver metastases upon systemic induction treatment: subgroup analysis of the CAIRO5 trial

BACKGROUND

RECIST may not be optimal for assessing treatment response with current systemic regimens. We evaluated RECIST, morphologic, and pathologically documented response (pathological response) in patients with initially unresectable colorectal cancer liver-only metastases (CRLM).

PATIENTS AND METHODS

Four hundred and eighty-nine patients from the phase III CAIRO5 trial were included who were treated with FOLFOX/FOLFIRI/FOLFOXIRI and bevacizumab or panitumumab. The association of the different response tools with overall survival (OS) was evaluated for all patients, and with early recurrence (<6 months) for patients after complete local treatment.

RESULTS

In the overall population, suboptimal [hazard ratio (HR) 1.10, 95% confidence interval (CI) 0.83-1.47] and optimal (HR 0.95, 95% CI 0.74-1.22) morphologic response were not associated with OS compared with no response. RECIST partial response (HR 0.61, 95% CI 0.49-0.76) and progressive disease (HR 5.77, 95% CI 3.97-8.39) were associated with OS compared with stable disease. In 242 patients who underwent local treatment, suboptimal (HR 1.22, 95% CI 0.76-1.96) and optimal (HR 1.28, 95% CI 0.89-1.86) morphologic response were not associated with OS compared with no response. RECIST partial response was not significantly associated with OS (HR 0.73, 95% CI 0.52-1.01), whereas progressive disease was (HR 19.74, 95% CI 5.75-67.78), compared with stable disease. While major pathological response (HR 0.66, 95% CI 0.44-0.99) was associated with OS, partial pathological response (HR 0.82, 95% CI 0.57-1.19) was not, compared with no pathological response. Pathological response, but not morphologic response and RECIST, was significantly associated with early recurrence (P < 0.001) which occurred in 13/58 (22%) patients with major response, 29/61 (48%) patients with partial response, and 51/88 (58%) patients with no response.

CONCLUSIONS

Our results show that RECIST but not morphologic response was prognostic for OS. In patients eligible for local treatment, neither RECIST nor morphologic response were associated with early recurrence. Pathological response was associated with early recurrence but is only available post-operatively. Hence, novel preoperative parameters are warranted to predict early recurrence and prevent potentially futile liver surgery.

Preoperative identification of early extrahepatic recurrence after hepatectomy for colorectal liver metastases: A machine learning approach

BACKGROUND

Machine learning (ML) may provide novel insights into data patterns and improve model prediction accuracy. The current study sought to develop and validate an ML model to predict early extra-hepatic recurrence (EEHR) among patients undergoing resection of colorectal liver metastasis (CRLM).

METHODS

Patients with CRLM who underwent curative-intent resection between 2000 and 2020 were identified from an international multi-institutional database. An eXtreme gradient boosting (XGBoost) model was developed to estimate the risk of EEHR, defined as extrahepatic recurrence within 12 months after hepatectomy, using clinicopathological factors. The relative importance of factors was determined using Shapley additive explanations (SHAP) values.

RESULTS

Among 1410 patients undergoing curative-intent resection, 131 (9.3%) patients experienced EEHR. Median OS among patients with and without EEHR was 35.4 months (interquartile range [IQR] 29.9-46.7) versus 120.5 months (IQR 97.2-134.0), respectively (p < 0.001). The ML predictive model had c-index values of 0.77 (95% CI, 0.72-0.81) and 0.77 (95% CI, 0.73-0.80) in the entire dataset and the validation data set with bootstrapping resamples, respectively. The SHAP algorithm demonstrated that T and N primary tumor categories, as well as tumor burden score were the three most important predictors of EEHR. An easy-to-use risk calculator for EEHR was developed and made available online at: https://junkawashima.shinyapps.io/EEHR/.

CONCLUSIONS

An easy-to-use online calculator was developed using ML to help clinicians predict the chance of EEHR after curative-intent resection for CRLM. This tool may help clinicians in decision-making related to treatment strategies for patients with CRLM.

RAS mutation associated with short surgically controllable period in colorectal liver metastases: a retrospective study

BACKGROUND

The prognostic implications of the RAS status in colorectal cancer liver metastasis (CRLM) remain unclear. This study investigated the prognostic significance of RAS status after curative hepatectomy, focusing on surgical controllability.

METHODS

This retrospective study included liver-only CRLM patients who underwent the first hepatectomy between 2015 and 2022 at the National Cancer Center Hospital. Recurrence-free survival (RFS), surgically controllable period (SCP), and overall survival (OS) were compared between RAS wild-type (RAS-wt) and mutant (RAS-mt) patients. Multivariate analyses were conducted to identify independent prognostic factors for each outcome and independent risk factors for less than 1 year SCP.

RESULTS

A total of 150 patients were evaluated, comprising 63 patients with RAS-mt status. There was no significant difference in RFS between RAS-mt and RAS-wt (7.00 vs. 8.03 months, p = 0.48). RAS-mt patients exhibited worse SCP (11.80 vs.21.13 months, p < 0.001) and OS (44.03 vs. 70.03 months, p < 0.001) compared to RAS-wt. Multivariate analysis identified RAS-mt as an independent prognostic factor for both OS (hazard ratio [HR]: 3.37, p < 0.001) and SCP (HR: 2.20, p < 0.001), and as an independent risk factor for less than 1 year of SCP (odds ratio, 2.31; p = 0.03).

CONCLUSIONS

CRLM with RAS mutations should be considered for strict surgical indications with preoperative chemotherapy and thorough examination, considering the possibility of short SCP.

A prediction model for response to immune checkpoint inhibition in advanced melanoma

Predicting who will benefit from treatment with immune checkpoint inhibition (ICI) in patients with advanced melanoma is challenging. We developed a multivariable prediction model for response to ICI, using routinely available clinical data including primary melanoma characteristics. We used a population-based cohort of 3525 patients with advanced cutaneous melanoma treated with anti-PD-1-based therapy. Our prediction model for predicting response within 6 months after ICI initiation was internally validated with bootstrap resampling. Performance evaluation included calibration, discrimination and internal-external cross-validation. Included patients received anti-PD-1 monotherapy (n = 2366) or ipilimumab plus nivolumab (n = 1159) in any treatment line. The model included serum lactate dehydrogenase, World Health Organization performance score, type and line of ICI, disease stage and time to first distant recurrence-all at start of ICI-, and location and type of primary melanoma, the presence of satellites and/or in-transit metastases at primary diagnosis and sex. The over-optimism adjusted area under the receiver operating characteristic was 0.66 (95% CI: 0.64-0.66). The range of predicted response probabilities was 7%-81%. Based on these probabilities, patients were categorized into quartiles. Compared to the lowest response quartile, patients in the highest quartile had a significantly longer median progression-free survival (20.0 vs 2.8 months; P < .001) and median overall survival (62.0 vs 8.0 months; P < .001). Our prediction model, based on routinely available clinical variables and primary melanoma characteristics, predicts response to ICI in patients with advanced melanoma and discriminates well between treated patients with a very good and very poor prognosis.

Clinical Prediction Models for Prognosis of Colorectal Liver Metastases: A Comprehensive Review of Regression-Based and Machine Learning Models

Colorectal liver metastasis (CRLM) is a disease entity that warrants special attention due to its high frequency and potential curability. Identification of "high-risk" patients is increasingly popular for risk stratification and personalization of the management pathway. Traditional regression-based methods have been used to derive prediction models for these patients, and lately, focus has shifted to artificial intelligence-based models, with employment of variable supervised and unsupervised techniques. Multiple endpoints, like overall survival (OS), disease-free survival (DFS) and development or recurrence of postoperative complications have all been used as outcomes in these studies. This review provides an extensive overview of available clinical prediction models focusing on the prognosis of CRLM and highlights the different predictor types incorporated in each model. An overview of the modelling strategies and the outcomes chosen is provided. Specific patient and treatment characteristics included in the models are discussed in detail. Model development and validation methods are presented and critically appraised, and model performance is assessed within a proposed framework.