Colorectal cancer (CRC), which is the third most common global malignancy, is a common malignant tumour threatening human health, and the annual incidence of CRC is approximately 1.2 million. The case fatality rate of CRC is ranked fourth, resulting in approximately 608 000 deaths per year.1 The development and progression of CRC is a multistep, multistage and polygenic process. In recent years, with the rapid development of the economic level in China and the change in dietary habits, the incidence of CRC has increased year by year, with new cases and deaths each year accounting for 30% of the same period CRC cases worldwide.2 Meanwhile, when found mostly in advanced stages, the prognosis is poor and the case fatality rate is high.3–5 Most CRCs develop from mutations in the APC gene in the common ‘adenoma-adenocarcinoma’ pathway; however, it has been shown that, in addition to adenomas, serrated polyps (SPs) also have a propensity to undergo carcinogenesis and approximately 20%–35% of CRCs develop through the ‘hyperplastic polyp-serrated adenoma-serrated adenocarcinoma’ pathway.6 It was reported that only 10%–15% of CRCs were derived from serrated adenomas in 2007, while by 2011, a study in America proposed that 35% of CRCs were derived from serrated adenomas.7 Previous studies on precancerous lesions have mainly focused on traditional adenomatous polyps and the ‘adenoma-adenocarcinoma pathway’, but recent studies on SPs have increased significantly. Increased awareness of SPs is important to prevent CRC.

SPs are a group of lesions in which the crypt epithelium exists with a serrated morphology and is heterogeneous, including hyperplastic polyps (HPs), sessile serrated adenomas/polyps (SSAs/Ps) and traditional serrated adenomas (TSAs).8 SPs are precancerous lesions that have been studied in recent years for their discovery of extreme malignant potential. According to the Consensus of China,9 advanced adenoma (AA) was defined as finding three or more adenomas or one of these adenomas measuring 10 mm or more in diameter, with one-third villous architecture or more, or with higher grade intraepithelial neoplasia in a colonoscopy. Multiple international guidelines and consensus agree that AAs have a higher risk of colorectal tumourigenesis and increased risk of metachronous advanced neoplasia.10 11

It is thought provoking that we found a significant number of patients presenting with both SPs and AAs synchronously. Then, some questions were asked: what are the clinical features of SPs with synchronous AA in people of different ages? Do SPs with synchronous AA significantly increase the risk of CRC and what is its pathogenesis? Although there are still relatively few studies on SPs with synchronous AA, their value is significant in the clinic. A detailed understanding of the clinical characteristics and carcinogenesis mechanism of SPs with synchronous AA can effectively block their progression and carcinogenesis risk, which plays a very important role in the prevention of CRC. To address the first question, we retrospectively analysed the clinical characteristics of SPs with synchronous AA in two cohorts of elderly and non-elderly individuals to provide a prospective basis for studies on the cooccurrence of SPs and AA.

In addition to adenomatous polyps, which have traditionally been considered precursors of CRC, the serrated adenoma pathway has been explored in the development of colorectal carcinogenesis.13 Previous epidemiological studies of SPs showed that the prevalence of SPs in the average risk population was 20%–40%, and most of them were HPs. A study in China found that14 the detection rate of SPs in all polyps increased year by year, reaching 8.74% by 2019. Among all SPs, 69.84% of them were diagnosed as HPs, 22.04% as SSAs/Ps and 7.76% as TSAs. This result is consistent with the findings of this study. TSAs and SSAs/Ps are considered precursors of CRC; therefore, they have an increased risk of synchronous and metachronous advanced tumours. The serrated pathway is characterised by mutations in the BRAF protooncogene, mutations in RAS and RAF, disruptions to the Wnt and P53 signalling pathways, the existence of microsatellite instability (MSI) and microsatellite stability, and widespread methylation of CpG islands. Complex internal correlations exist among each pathway, but they also show differences. SPs and traditional AAs can occur simultaneously, although they originate from two different biological pathways. It should be noted that current guidelines do not address this combination of pathologies. People with simultaneous SPs and AAs have been proven to be a high-risk population for CRC. According to a study assessing the risk after SSAs/Ps with or without synchronous adenoma compared with that following a conventional adenoma, synchronous conventional adenomas may increase the malignant potential of SSA/P. This study revealed that high-risk adenoma (HRA) (HR=2.04, 95% CI 1.70 to 2.45) and SSA/P with synchronous adenoma (HR=3.20, 95% CI 1.31 to 7.82) were independent predictors of future advanced neoplasia compared with low-risk adenoma.15 Another study demonstrated that the cumulative incidence of 3-year isochronous advanced CRC in patients with both SPs and adenoma was 17.9% (95% CI 8.0% to 27.6%), which was significantly higher than 10.7% (95% CI 7.7% to 3.6%) in patients with simple adenoma.16 The coexistence of SPs and AAs significantly increased the risk of heterochrony advanced CRC, and these people may have different risk factor characteristics than patients with only SPs or AAs. This study, therefore, analysed the clinical characteristics of SPs with synchronous AA in elderly and non-elderly people in detail.

A prospective analysis revealed that independent predictors of developing CRC in patients who had SPs with AA were age ≥65 years old (OR=1.15, 95% CI 1.05 to 1.26) and male sex (OR=2.13, 95% CI 1.3 to 3.49),17 while our study led to the same conclusion that the proportion of SPs with synchronous AA was higher in elderly or male patients. In another prospective study,18 30.5% of patients with serrated polyposis syndrome (SPS) were diagnosed with synchronous AA/CRC. The findings suggested that older age and higher BMI were independent risk factors for SPS, which was consistent with the findings in our study that patients who had SPs with high BMI were prone to the development of comorbid AA in the elderly group. The age-related mechanism of SPs is complex,19–21 involving the following aspects: (1) Chronic inflammation of the gut: With ageing, the intestinal mucosa may be more susceptible to chronic inflammation, and prolonged inflammatory stimulation may lead to cell proliferation and polyp formation. Intestinal inflammation may lead to repeated cell damage and repair, increasing the incidence of these molecular events (BRAF mutation, KRAS mutation, CpG island methylation and MSI), which may promote the development and carcinogenesis of SPs. (2) Alterations in molecular biology: SPs is characterised by genetic and epigenetic alterations that may be more common in elder adults because of the increased likelihood of mutation accumulation due to the increased number of cell divisions that occur with age. (3) Immunosenescence: Patients with SPs experience immunosenescence and decreased immune surveillance function. The function of the immune system in the elderly gradually declines with age, a phenomenon known as immunosenescence. Immunosenescence can lead to the weakened ability of immune cells to monitor and clear tumours, which may increase the risk of SPs occurrence and malignant transformation.

Current smoking is a common high-risk factor for both SPs and AAs. Studies have found that current smoking increases the risk of SPs or HRAs by nearly threefold, and it increases the risk of SPs plus HRA by more than eightfold.22 A systematic review and meta-analysis found that alcohol intake (RR 1.33; 95% CI 1.17 to 1.52) and BMI (RR 1.40; 95% CI 1.22 to 1.61) were also risk factors significantly increasing SPs, in addition to smoking (RR 2.47; 95% CI 2.12 to 2.87), and SSA/P was directly associated with smoking and drinking compared with HP, except for BMI.23 Previous studies have also suggested risk factors for SSA/P, including Caucasian ethnicity, family history of CRC, smoking and drinking.24 These findings enhance our understanding of the mechanisms underlying the development of SPs and suggest that the risks of CRC from the serrated pathway could be reduced with lifestyle changes. When data from large prospective studies were analysed,25 researchers reported that smoking, BMI, drinking, family history of CRC and height were associated with a higher risk of SPs and conventional adenomas, respectively, and this association was stronger, especially in synchronous SPs and conventional adenomas. However, smoking, BMI and drinking were more closely associated with SPs than with conventional adenomas, and our study found that all three associated risk factors were also strongly associated with SPs with synchronous AA in both the non-elderly and elderly groups. These findings provide support for the aetiological heterogeneity of colorectal neoplasia. It has been shown26 that independent risk factors for TSA include age ≥50 years old, hypertension and current smoking, with 59.5% of patients with TSAs synchronously having AAs. AAs and HRAs can be more frequently observed in patients with TSAs than in patients with adenomas alone. Another study also showed27 that patients with one or more TSAs had a sevenfold higher risk of developing synchronous advanced CRC than those with one or more HPs. In our study, a higher proportion of TSAs with synchronous AA was found; however, it did not differ statistically between the elderly and non-elderly groups.

In general, we believe that all SPs are considered to be precursors of CRC, except HPs. HP is the most common type of SP, and its incidence is approximately 4.88%.28 An HP with a diameter ≤5 mm, which is mostly distributed in the distal colon and rectum,29 has the same distribution pattern as a TSA.30 Some previous studies have suggested that some HPs might be associated with an increased risk of CRC. Nevertheless, a prospective study on the risk of adenoma recurrence associated with HPs suggested that patients with coexistent HPs and adenomas were not found to have an increased risk of adenoma recurrence in the subsequent 3-year follow-up evaluation, regardless of HPs located in the proximal or distal colon.31 In our study, patients with HPs ≥6 mm and morphological types Isp, Ip and IIb were found to be more likely to have AAs both in the elderly and non-elderly groups. We also found that HPs in the non-elderly group were more likely to have synchronous AA when they occurred in the proximal colon. For patients who have HPs with synchronous AA, whether elderly or non-elderly, we should treat them carefully according to the principles of management and surveillance for AAs.

The incidence of SSA/P is lower than that of HP, approximately 1.11%, and is mainly located in the proximal colon. The incidence rate of TSA is approximately 0.66%, and TSA is mainly located in the distal colon.28 Colonoscopy showed that the prevalence of SSA/P and TSA was influenced by patients, endoscopic technique and pathological diagnosis. The prevalence of SSA/P was approximately dozens of times that of TSA, while the malignant potential of SSA/P and TSA was comparable.32 SSA/P, usually located in the proximal colon and with a diameter of >10 mm, is a sessile flat eminence with a pale surface that is similar to the colour of the mucosal background, which is difficult to find under ordinary endoscopy and can easily cause a missed diagnosis.33 In addition to being prone to occur in the proximal colon (ascending colon and ileocecal region), SSAs/Ps with synchronous AA in the non-elderly were also prone to occur in the rectum in this study. Studies suggest that a high-fat diet, red meat, obesity and smoking are high-risk factors for left colon and rectal lesions,34 and the incidence of SSA/P in the left colon and rectum is increasing as people’s living standards improve and dietary habits change. Age ≥65 years old is an independent predictor of the development of CRC from AAs or SSAs/Ps17; however, this study found that the proportion of SSAs/Ps with synchronous AA in the non-elderly group was significantly higher than that in the elderly group, suggesting that CRC should also be monitored in non-elderly patients with SSAs/Ps. Yang et al found35 that three features on white light endoscopy, including a mucinous cap, indistinct border and cumulus-like surface structure, were independent factors for predicting SSA/P. Tadepalli et al pointed out36 that a mucinous cap was the most common feature of SSA/P. In this study, the above characteristics were confirmed and are also an important basis for the diagnosis of SSA/P by colonoscopy. Some scholars demonstrated that SSAs/Ps were larger and more likely to be accompanied by synchronous and metachronous advanced CRC (6.79% and 6.08%, respectively) than HPs (2.14%). They found that large SPs (diameter ≥10 mm, OR: 2.52, 95% Cl 1.40 to 4.55) and SSAs/Ps with high-grade intraepithelial neoplasia (OR 13.85, 95% CI 3.28 to 58.56) were associated with an increased risk of synchronous advanced CRC.14 Traditionally, the larger the SSA/P volume is, the more likely it is to be accompanied by dysplasia or cancer. In this study, univariate analysis suggested that synchronous AA was closely associated with SSA/P diameter ≥10 mm in the non-elderly group, whereas it was easily associated with SSA/P diameter ≥16 mm in the elderly group. Those SSA/P with morphology types Ip and IIb were more likely to have synchronous AA. Multivariate analysis in this study also identified SSA/P diameter ≥16 mm as a risk factor for synchronous AAs in the non-elderly group. Thus, it is suggested that we should be alert to the risk of synchronous AA and metachronous advanced neoplasia, in addition to the risk of synchronous advanced neoplasia when finding a larger SSA/P of types Ip and IIb during colonoscopy. These patients with SSAs/Ps may be high risk for the development of CRC in the future and a population for focused follow-up surveillance.

Current resection criteria for SPs are not uniform. American researchers37 recommend that all serrated lesions in the proximal sigmoid colon and those serrated lesions >5 mm in diameter in the sigmoid colon and rectum be completely resected. Japanese scholars38 suggest that endoscopic resection for SSAs/Ps ≥10 mm and TSAs ≥5 mm be performed, and HPs ≥10 mm located in the proximal colon also be treated with endoscopic resection, which is difficult to distinguish from SSAs/Ps. However, how to define the resection standard of SPs needs further research and verification for SPs with synchronous AA.

This study also has some limitations. First, the current study is a retrospective study. Second, the current study was conducted in three tertiary endoscopy centre in one city, so selection bias cannot be ignored. Third, because most of the subjects in our study were symptomatic patients who underwent colonoscopy to clarify the cause and a small number were asymptomatic patients who underwent colonoscopy for health screening, there may be bias in colonoscopy data. We will conduct prospective studies to continuously observe the morbidity characteristics, treatment and clinical outcomes of SPs with synchronous AA. Fourth, AA criteria vary across countries, each criterion does not carry the same risk for CRC. Because our study is the analysis of clinical characteristics and risk factors on SPs with synchronous AA in the Chinese population. Therefore, we have applied the diagnostic criteria of AA in the Chinese consensus. In the future, we will expand the scope of our research, draw on the international guidelines and make our research more scientific and universal. Fifth, drinking history may not carry the same effect across different populations. In the future, our research will be more detailed and quantified. Drinking history will be precisely defined as daily alcohol consumption and daily drinking frequency. Additionally, the exclusion criteria omitted some high-risk patients, such as those with inflammatory bowel disease or hereditary polyposis syndromes, which may limit the applicability of the findings to these populations.

In conclusion, this study showed that the incidence of HPs with synchronous AA in the elderly group (312, 35.06%) was significantly higher than that in the non-elderly group (695, 30.56%), while the incidence of SSAs/Ps with synchronous AA in the non-elderly group (196, 56.16%) was remarkably higher than that in the elderly group (80, 44.20%). Multivariate analysis revealed that male sex (OR 0.798, 95% CI 0.653 to 0.975), drinking history (OR 1.481, 95% CI 1.224 to 1.791) and HP size (≤20 mm, ORs range from 1.885 to 5.492, 95%CIs range between 1.513 to 2.347 and 3.046 to 9.902) were independent facilitating factors for HPs with synchronous AA in the non-elderly group, while drinking history (OR 1.910, 95% CI 1.436 to 2.540) and HP size (≤15 mm, ORs range from 1.845 to 1.964, 95% CIs range between 1.316 to 2.588 and 1.306 to 2.953) were independent facilitating factors for HPs with synchronous AA in the elderly group. For SSAs/Ps with synchronous AA, male sex (OR 0.381, 95% CI 0.208 to 0.699), smoking history (OR 0.505, 95% CI 0.295 to 0.863), drinking history (OR 0.512, 95% CI 0.301 to 0.868) and SSA size (≥16 mm, ORs range from 8.338 to 11.689, 95% CIs range between 1.057 to 65.743 and 1.680 to 81.328) were independent facilitating factors in the non-elderly group; nevertheless, only higher BMI (OR 2.230, 95% CI 1.127 to 4.411) was an independent risk factor in the elderly group. This study has important significance for clarifying the clinical characteristics and risk factors for SPs with synchronous AA in different age groups. It also provides a theoretical basis for the development of a digestive endoscopy screening plan or follow-up period after colorectal polypectomy for different risk groups to improve the early detection and early endoscopic treatment of SPs and AAs.