In this study, we investigated the potential impact of pre-operative elevated lipid levels on the clinical outcomes OS and DFS, in patients with CRC who underwent curative-intended surgery for stage I-III CRC.
The main analysis, after adjusting for propensity scores, did not demonstrate a significant association between pre-operative TC levels above 4 mmol/L and OS in patients with CRC. Similarly, no significant association was found between elevated TC levels and DFS. However, in the first subgroup analysis, focusing on patients with both TC above 4 mmol/L and LDL above 3 mmol/L, a significantly higher OS was observed in the target group compared to the control group. No significant associations were found in the other subgroup analyses, apart from improved OS in patients with UICC I-II and TC > 4 mmol/L as well as improved DFS in patients with exposure to statins as well as TC > 4 mmol/L. Interestingly, we found a statistically significant association between improved OS and DFS in patients with TC > 4 mmol/L in the unadjusted analysis for both outcomes.
These findings suggest that while pre-operative TC levels alone may not be strong prognostic markers of survival in patients with CRC, there might be an interaction between TC and LDL levels that is associated with OS, as well as an observed benefit on DFS through the exposure to statins.
Our findings suggest a potential paradoxical association between higher cholesterol levels and overall survival. One possible explanation for our observed association is that elevated cholesterol levels may serve as a marker of better nutritional status in cancer patients. Malnutrition and cachexia are common complications of cancer and are associated with unfavorable outcomes [29]. Patients with higher cholesterol levels might possess a better ability to maintain their nutritional status and preserve muscle mass, which could contribute to improved cancer outcomes. A study conducted to investigate the metabolic changes that occur during cancer cachexia progression finds that both cholesterol and LDL, but not triglycerides or HDL, were significantly reduced in cachectic patients [30]. This underlines previous findings as well [31]. However, this hypothesis warrants further investigation and consideration in future studies. Furthermore, it's important to acknowledge that the data used in this study only encompasses patients having undergone surgery with curative intent up until May 2020, thus part of the follow-up period for patients undergoing treatment for colorectal cancer in 2020 may have overlapped with the COVID-19 pandemic.
Emerging evidence suggests that lipid disturbances during COVID-19 might also have contributed to disease severity and complications [9–11]. Dyslipidemia could potentially worsen the immune response and exacerbate the inflammatory processes associated with severe COVID-19 cases. Given this emerging evidence, it would be intriguing to investigate in future research whether the pandemic has impacted the observed relationship between lipid metabolism and clinical outcomes in colorectal cancer patients. The confluence of dyslipidemia, immune dysfunction, and inflammation, observed in both cancer and viral diseases like COVID-19, underscores the relevance of understanding lipid metabolism in diverse disease contexts.
The strengths of this study lie in the utilization of a large number of variables in the propensity-score analysis (42,190 in total) which enabled the computing of detailed propensity scores. This approach creates a real-world setting that attempts to simulate the gold standard of a randomized controlled trial (RCT) [26]. Another strength in using data from national Danish registries is the uniform data collection, as it is standardized across regions and covers the entire population. This way, selection- and attrition bias is close to absent as there would only be loss to follow-up if patients leave the country. Health seeking bias has been accounted for in the analysis examining patients who have had their lipids measured versus the ones who have not. Furthermore, using the open-source tools provided by the OHDSI community provides a framework for observational health studies that ensure standardized analytics through the use of templates. This improves reproducibility and transparency while ensuring consistency in how data is computed.
However, while the PS matching controls for measured confounders, many covariates that would be valuable to consider when computing the propensity scores are not included in the CDM, such as socioeconomic status, with a risk of residual confounding remaining. Regarding pre-analytical conditions, it is essential to note that information regarding patient fasting status prior to LDL analysis was not available. As outlined in the literature [32], fasting before sample collection is recommended for certain analyses, including LDL measurement. However, the lack of data on preceding fasting necessitates cautious interpretation of our results, as fasting status can significantly influence LDL levels.
These limitations introduce inherent biases and constraints in data collection. Therefore, further studies with larger sample sizes and diverse patient populations are necessary to confirm our findings and elucidate the underlying mechanisms explaining the unexpected association between higher cholesterol levels and longer overall survival in colorectal cancer.
In conclusion, this paper explores the question of cholesterol's impact on survival in surgery-treated patients with colorectal cancer, suggesting that, contrary to expectations, higher total cholesterol and low-density lipoprotein levels are associated with improved overall survival, but not elevated TC levels alone. This finding challenges established knowledge and calls for further research to elucidate the complex interactions between cholesterol metabolism, nutritional status, and cancer outcomes. Understanding these mechanisms holds promise for refining risk assessment strategies and potentially influencing patient care and outcomes.