1. This prospective study explores the association between DRRD score and the incidence of lung cancer for the first time in a large population, with a long follow-up. The results encourage people to adhere to DRRD score in their daily diet.

2. We interestingly found that the inverse association between DRRD score and the incidence of lung cancer was more pronounced in participants with a clear family history of lung cancer, although the reason is still exactly unclear.

Lung cancer is the second most diagnosed malignancy worldwide and new cases of lung cancer accounted for 11.4% of all cancers in 2020 (1). It is presented as the leading cause of all cancer deaths, with an estimated 1.8 million deaths each year (1, 2). The association between diet and lung cancer has been studied by researchers globally and some conclusions have been subsequently reached. Evidence suggested specific dietary habits might influence lung cancer risk and play a role in lung cancer prevention. Dietary inflammatory index (DII), calculated to assess the inflammatory levels in one’s diet, has been implicated in the pathogenesis of lung cancer (3). In a population-based, prospective cohort followed up for 17.5 years, the risk of lung cancer was decreased by 10% in the highest quality group compared with the lowest one based on quintiles for DII (4). In an updated pooled analysis including five cohort studies (n = 12,730 incident cases), higher adherence to the Mediterranean diet, a plant-based pattern, has been reported to decrease the risk of lung cancer by 16% (5). Both patterns encouraged good diet quality to reduce inflammation levels in the human body (6, 7). In recent years, a dietary pattern for diabetes prevention named diabetes risk reduction diet (DRRD) was established (8). Nine dietary variables including cereal fiber, nuts, coffee, whole fruits, the ratio of polyunsaturated to saturated fat, glycemic index (GI), trans-fat, sugar-sweetened beverages (SSBs)/fruit juices, and red and processed meats composed of DRRD (8). Higher DRRD score was suggested to have the potential to reduce the possibility of hyperinsulinemia and insulin resistance (9), which were demonstrated to be closely related to increased cancer risk, including lung cancer (10–12). The mechanism may be associated with the activation of insulin-like growth factor-1, stimulating of the Ras signaling pathway, local angiogenesis or growth promotion via insulin receptors present on lung cancer cells (11, 13, 14). Therefore, we hypothesize that higher DRRD score may be inversely associated with lung cancer risk (15). This study aims to evaluate the association between DRRD score and the risk of lung cancer in a large population.

This study explores the association between DRRD score and lung cancer in a large population. The results show an inverse association between DRRD and the risk of lung cancer, even after adjusting for confounding factors. Though DRRD score has been developed to prevent T2D, the significant inverse association was detected among the American population. The dose-response analyses also present a declining linear tendency of the risk of lung cancer with DRRD score, indicating the risk of lung cancer may change in a parallel manner with the change of DRRD score. The sensitivity analysis suggests the significant relationship between DRRD score and the incidence of lung cancer is robust by excluding outliers.

Previous studies indicated participants with diabetes are more prone to develop cancers (12, 20) and insulin resistance has been suggested to be associated with increased lung cancer risk (10, 11). Hyperinsulinemia and hyperglycemia may accelerate the biological aging progress (12, 15) and stimulate cellular signaling pathways associated with growth factor-dependent cell proliferation and cancer development (21, 22). Moreover, insulin increases the activity of insulin-like growth factor-1 involved in tumor initiation and progression (23, 24). These processes are likely affected by the DRRD score, which has been suggested to be inversely associated with not only T2D risk but also cancers (8, 25, 26). In addition, higher intake of fiber (27, 28), nuts (29), coffee (30, 31), polyunsaturated fat (32), and fruits (33) are associated with a lower level of inflammation (29, 34, 35), but high GI diet (36), trans fatty acids (37), SSBs (38), and red and processed meats (39) are positively associated with inflammation (40, 41). Namely, all the component parts of DRRD are closely related to chronic inflammation, which is also involved in tumorigenesis (12, 15). We hypothesize the inverse association between DRRD score and the incidence of lung cancer is possibly because of the ease of chronic inflammation, hyperinsulinemia, and insulin resistance. Here, people are encouraged to adopt a healthier dietary habit and have higher intake of cereal fiber, nuts, coffee, fruits, and polyunsaturated fat, which indicates a higher DRRD score; but have less intake of a high GI diet, trans-fat, SSBs/fruit juices, saturated fat and red, processed meats, which indicates a lower DRRD score.

Interestingly, we found an even stronger inverse association between DRRD score and the incidence of squamous cell carcinoma, but not the incidence of adenocarcinoma, large cell carcinoma, or small cell carcinoma. Previous studies also suggested the inverse associations with dietary fiber, yogurt, fruits, and vegetables consumption, even some dietary habit index including Alternative Healthy Eating Index-2010, the alternate Mediterranean Diet score, the Dietary Approaches to Stop Hypertension score, and the Dietary Inflammatory Index were more evident for squamous cell carcinoma than other histological cell types (4, 42). The reason is not exactly clear. In a previous study, among the main histological types of lung cancer, squamous cell carcinoma is the most strongly associated with smoking, which increases the risk of lung cancer in part through its pro-oxidant properties (43). We hypothesize that the strong association between DRRD and the incidence of squamous cell carcinoma is related to the mitigation of chronic inflammation. More research should be conducted to detect the reasons behind this.

In subgroup analyses, the magnitude of associations did not change appreciably stratified by age, sex, BMI, race/ethnicity, smoking status, or history of diabetes. However, the significant association between DRRD score with the incidence of lung cancer was more pronounced in participants who had a clear family history of lung cancer, with an adjusted HR of 0.62 (95% CI 0.44-0.87). Consistently, dietary factors were previously found to be associated with cancer risk among individuals with a family history of certain cancers (44, 45). Previous evidence suggests gene expression may be changed by dietary factors via epigenetic mechanisms (46). For instance, the consumption of a ‘‘Western-type’’ diet was found to functionally alter the hepatic gene expression by affecting histone polyacetylation and reducing short-chain fatty acids (47). Similarly, the result of our stratified analysis in family history of lung cancer could be attributed to the heritability of genes related to lung cancer susceptibility. The susceptible genes of individuals with a family history of lung cancer may be regulated by dietary habits in the long run. Further research is needed to explore whether the association between DRRD score and lung cancer risk could be explained by the regulation of genes related to both diabetes and lung cancer. The subgroup analyses also showed the opposite association between DRRD and lung cancer risk in former/current smokers and nonsmokers. Because smoking increases the level of inflammation in the human body and thus raises the risk of lung cancer (48), the inverse association between DRRD and lung cancer might be more evident in smokers since adherence to DRRD indicated the alleviation of inflammation.

Of note, this study has remarkable strengths. This prospective large-scale study explores the association between DRRD score and the incidence of lung cancer for the first time in a large population. In this study, the follow-up time was calculated from the completion of DHQ. Because the average time from randomization to the completion of DHQ was approximately 3 years, the actual time for observation in this cohort was far more than 8 years. An appropriate observation time ensured the event to be obtained. In addition, the results were adjusted for a wide range of potential confounding factors, though we could not exclude the possibility that more unmeasured residual confounders might influence the observed association. Furthermore, the sensitivity analyses were conducted to assess the robustness of the association between DRRD score and lung cancer. Moreover, we interestingly found the inverse association between DRRD score and the incidence of lung cancer was more pronounced in participants with a clear family history of lung cancer, though the reason is still unclear exactly.

However, shortages exist in this study. First, the dietary habits may change during the long follow-up, but using a baseline diet only to evaluate the dietary intake generally yielded weaker associations with the incidence of disease than using the cumulative dietary intake (49). Second, the information of self-reported DHQ may not be precise enough because of the huge contents. Third, some subgroups included too small numbers of outcome events to allow sufficient statistical power to observe significant associations and detect potential interactions between DRRD score and stratified factors. Fourth, there were around 36.6% participants (56,728 of 154,887) excluded because they had invalid BQ or DHQ, or were diagnosed with cancer before completing the DHQ. Selection bias might exist in this process though the baseline information could not be compared between the excluded group and included one due to several missing values.

In conclusion, this study shows an inverse association between DRRD score and risk of lung cancer. People are encouraged to have more cereal fiber, nuts, coffee, fruits, diet with higher ratio of polyunsaturated to saturated fat, which indicates higher DRRD score; but have less intake of high GI diet, trans-fat, SSBs/fruit juices, and red and processed meats, which indicates a lower DRRD score. Further studies should be scheduled to confirm the association.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.

This study has been approved by the United States NCI (CDAS project “PLCO-800”). The written informed consent to participate in the study was provided by each participant, and the study protocol was approved by the Institutional Review Board of the United States NCI (https://biometry.nci.nih.gov/cdas/plco/). The participants provided their written informed consent to participate in this study.

YZ: Conceptualization, methodology, software, formal analysis, data curation, writing - original draft. GZ: Conceptualization, methodology. MZ: Visualization, investigation, validation. LC: Visualization, investigation. HW: Writing - review and editing, supervision. FL: Writing - review and editing, supervision. All authors contributed to the article and approved the submitted version.

This work was supported by National Natural Science Foundation of China grant (NSFC No.32070764; No.81800087); 1.3.5 project for disciplines of excellence, West China Hospital, Sichuan University (ZYJC18021); Sichuan Science and Technology Program (No.2021YFQ0030).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.