Introduction

Colorectal cancer (CRC) ranks as the most common cancers globally, placing third and second in terms of prevalence and mortality, respectively [1,2,3]. Reflecting this global trend, the prevalence of CRC is increasing in Western and Asian countries [4]. While the historical incidence rate of CRC in Iran was lower compared to Western countries [5], it has been continuously increasing in recent years [5]. CRC now ranks as the fifth most prevalent cancer among Iranian men and the third most prevalent among Iranian women [6]. This highlights the growing public health burden of CRC in Iran, with over 90,000 new cancer cases diagnosed annually [7].

Diet is a recognized and influential factor in CRC occurrence, with the potential for both harmful and protective effects on its prevalence [8]. A plant-based diet, in particular, has been shown to modulate several CRC risk factors, potentially leading to a reduced risk of developing the disease [9]. Fruits and vegetables are rich in fiber, folate, vitamins, and antioxidant compounds, which have been linked to a protective effect against CRC [10]. In contrast to refined grains, whole grains contain the germ and bran, rich sources of fiber, antioxidants, and potentially other anticarcinogenic phytochemicals [11].

Several observational studies have reported an association between adherence to a plant-based diet and reduced risk of CRC. For instance, one study showed following a plant-based diet was linked to a 46% reduced risk of colon cancer and a 73% reduced risk of rectal cancer [12]. Similarly, another study noted a decrease in the risk of CRC with a plant-based dietary pattern [13].

Recently, novel measures of plant-based diets have been developed (plant-based diet index (PDI), unhealthful PDI (uPDI), and healthful PDI (hPDI)), to evaluate consumption of both animal and plant-based foods, while also considering the quality of plant-based options [14].

While limited research has investigated the association between these indices and CRC risk [14, 15], to our knowledge, no studies has explored this correlation within Middle Eastern populations. Notably, the dietary habits of Middle Eastern population differ significantly, with large portion sizes and a preference for refined grains like bread and white rice, leading to a higher carbohydrates intake compared to other regions [16]. Therefore, this study aimed to investigate the association between PDI, uPDI, hPDI, and the risk of CRC in Iranian adults, contribution to a deeper understanding of the link between diet and CRC risk.

Materials and methods

Study population

This hospital-based case-control study was conducted in three general hospitals and their associated CRC surgery departments of Tehran, Iran. The sample size was calculated considering a β = 0.2, α = 0.05, and anticipated odds ratio (OR) of 0.45 based on a previous study investigating the association between a similar dietary factor and CRC risk [17].

Cases were individuals aged 40 to 75 years who were diagnosed with CRC, and had no prior history of cancer by convenient sampling method. The control group was randomly selected from the same hospitals, and free of chronic diseases or acute illness that could influence dietary pattern. The diseases of the patients in the control group include 14.1% osteoarticular disorders, 38% sprains and fractures, 11.3% disk disorders, 9.8% acute surgical conditions, 7.0% trauma, 7.0% skin diseases, and 12.0% other diseases.

Each CRC patient was paired with two control patients. A total of 267 patients were selected (89 cases and 178 controls). Twenty-four patients did not participate, and 30 were excluded due to dietary reporting error (total energy intake outside the defined range of ± 3 standard deviations from the mean or incomplete food frequency questionnaire (FFQ)). Statistical analysis was performed on the remaining 71 cases and 142 controls. This study received approval from the Medical Research and Ethics Committee of Shiraz University of Medical Sciences (IR.SUMS.SCHEANUT.REC.1401.011). Some details of this study have been published previously [18,19,20].

Socio-demographic and anthropometric evaluations were conducted

Skilled interviewers administered valid questionnaires to all participants, collecting data on socio-demographic characteristics, family history of CRC, typical vegetable consumption methods, physical activity levels, medication use, smoking habits, and dietary intake. A nutritionist assessed participants’ body measurements (weight with a precision of 0.1 kg; for hospitalized patients, weight at admission was used).

The dietary assessment included the calculation of plant-based diet scores

To assess usual food intake, a face-to-face interview was conducted with all participants using a validated 168-item FFQ [21]. Food intake data was then converted to grams per day using Nutritionist IV software (version 7.0; N-Squared Computing, Salem, OR, USA). This software computed mean energy intake and nutrient intake for all participants.

The Satije et al. method [22,23,24] was employed to create three variants of the plant-based diet index: PDI, uPDI, and hPDI. The FFQ was classified into 18 food groups within three main categories: animal foods (dairy, egg, meat, seafood, animal fat, and other animal-based foods), healthy plant foods (whole grains, nuts, vegetables, fruits, tea or coffee, vegetable oils, and legumes), and unhealthy plant foods (sugar-sweetened beverages, potatoes, sweets or desserts, refined grains, and fruit juices). The scoring system assigned a score of 10 to the highest consumption and 1 to the lowest consumption for PDI and hPDI. Conversely, uPDI received a score of 1 was given for the highest consumption and 10 for the lowest consumption. The final score for each index ranged from 18 to 180, with a higher total score indicating greater adherence to that specific plant-based dietary pattern.

Statistical analysis

Data were analyzed using SPSS software (version 23.0), SPSS Inc, Chicago IL. The normality of the data was assessed using the Kolmogorov-Smirnov test. Quantitative variables were compared between the case and control groups using either the Mann-Whitney U test or the independent samples t-test, while qualitative variables were compared using the chi-square test. To evaluate the association between PDI, hPDI, and uPDI scores and CRC, both crude and adjusted logistic regression models were utilized (adjusted for confounder variables: physical activity, smoking, common method of vegetable consumption, energy and fiber intake, family history of CRC, taking the mineral supplement, acetaminophen, and aspirin). Also, all figures were depicted by R software.

Results

The baseline characteristics of the study participants are presented in Table 1. Notably, cases had significantly higher uPDI scores compared to controls (p = 0.003). Conversely, hPDI score was significantly more in the control group (p = 0.001). Cases were less likely to report a. Fiber intake (p˂0.001), family history of CRC (p = 0.017), use of mineral supplements (p = 0.015), aspirin (p = 0.016), and acetaminophen (p = 0.004) also showed significant differences between the groups.

Table 1 Baseline features of the study population
Full size table

The participants’ food group intakes are shown in Figs. 1, 2 and 3. The participants’ food group intakes based on PDI tertiles are reported in Fig. 1. In the last tertile of PDI, the mean intake of whole grains (p˂0.001), fruits (p˂0.001), vegetables (p˂0.001), nuts (p˂0.001), legumes (p = 0.001), vegetable oils (p = 0.002), tea and coffee (p˂0.001), refined grains (p = 0.015), potatoes (p˂0.001), and sweets and desserts (p = 0.001) were significantly more.

Fig. 1
figure 1

Participants’ food group intake across tertiles of PDI

Full size image
Fig. 2
figure 2

Participants’ food group intake across tertiles of hPDI

Full size image
Fig. 3
figure 3

Participants’ food group intake across tertiles of uPDI

Full size image

The participants’ food group intakes based on hPDI tertiles are reported in Fig. 2. The mean intake of refined grains (p = 0.009), potatoes (p = 0.042), sugar and sweetened beverages (p˂0.001), sweets and desserts (p˂0.001), animal fat (p˂0.001), eggs (p = 0.024), fish and seafoods (p˂0.001), meats (p˂0.001) and animal based foods (p˂0.001) were significantly lower, but vegetables (p = 0.002), intake was more in the last tertile of hPDI.

The participants’ food group intakes based on uPDI tertiles are reported in Fig. 3. The mean intake of In the last tertile of uPDI, the mean intake of whole grains (p˂0.001), fruits (p˂0.001), vegetables (p˂0.001), nuts (p = 0.006), legumes (p = 0.004), vegetables oils (p = 0.002), animal fat (p = 0.003), dairy products (p˂0.001), eggs (p = 0.012), fish and sea foods (p˂0.001) and meats (p˂0.001) were significantly lower in the last tertile of uPDI.

Table 2 presents the crude and multivariable-adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for CRC risk according to tertiles of PDI, hPDI, and uPDI scores. No significant association was observed between PDI and CRC (Ptrend=0.609). In contrast, higher adherence to the hPDI was associated with a significantly lower risk of CRC in the highest tertile (79% reduction) compared to the lowest tertile. There was also a non-significant trend towards a lower risk in the second tertile (43% reduction). For the uPDI, participants in the second and highest tertiles had a higher risk of CRC compared to those in the first tertile (tertile (T)2: OR = 2.99; 95% CI: 1.12–7.96 - and T3: OR = 6.76; 95% CI: 2.41–18.94). No significant association was found between the highest tertile of PDI and CRC risk compared to the lowest tertile.

Table 2 Association between PDI, hPDI, and uPDI with CRC
Full size table

Discussion

Our study suggests that following a healthy plant-based diet (higher hPDI scores) may be associated with a lower risk of CRC, while a diet high in unhealthy plant-based diet (higher uPDI scores) might be linked to an increased risk. No significant association was found with overall plant-based diet index (PDI).

The current study’s findings demonstrated an inverse relationship between higher hPDI tertiles and CRC risk. The aligns with previous research by Kim et al. [14] who reported a 20% reduction in CRC risk. Additionally, a study by Wang et al. [15] indicated that higher hPDI scores were associated with a reduced risk of CRC. These findings collectively support the potential protective role of healthy plant-based dietary patterns against CRC.

The inverse association between high hPDI scores and CRC risk observed in the present study and previous research can be explained by several potential mechanisms. Notably, hPDI emphasizes consuming a greater quantity of nutrient-dense plant-based foods, rich in fiber and antioxidants [25]. In contract, red meat consumption has been linked to an increased risk of CRC [26]. A meta-analysis study suggests that a daily increase of 150 g of red meat could increase elevate CRC risk by 20% [27]. The potential mechanisms for red meats carcinogenicity include the generation of N-nitroso compounds, heme iron, sulfur-containing amino acids, and saturated fatty acids [28]. Furthermore, fermentation of fiber in the colon by gut microbiota products short-chain fatty acids, such as butyrate, which can induce apoptosis and inhibit cancer cell proliferation. Additionally, dietary antioxidants neutralize free radicals that cause DNA damage and impede cancer cell formation [29,30,31].

A healthy gut microbiome, compared to a disrupted one, is associated with various health benefits, and a strong link has been established between its composition and cancer risk, particularly CRC [32, 33]. Certain plant foods rich in fructo-oligosaccharides and galacto–oligosaccharides (e.g., onions, garlic, artichokes) promote the growth of beneficial bacteria that can reduce CRC risk through multiple mechanism, including preventing chronic inflammation, protecting intestinal cells, improving insulin sensitivity, and promoting healthy cell division [34].

Furthermore, dietary polyphenols, such as Flavonols and anthocyanins, found abundantly in plant-based foods characteristics of high hPDI scores, can improve gut microflora composition [35, 36]. The cyclooxygenase-2 (COX-2) pathway plays a role in inflammation and cancer development. Its activation, induced by inflammatory cytokines like interleukin-6 (IL-6), can lead to cell proliferation, angiogenesis, and other changes that promote tumorigenesis. Conversely, anti-inflammatory diets, characterized by high hPDI scores, have been shown to suppress COX-2 pathway activation [31]. This suggests that hPDI may exert protective effects against CRC by modulating inflammatory processes. In conclusion, study suggests that adherence to an anti-inflammatory dietary pattern, reflected by high hPDI scores, rich in antioxidants, polyphenols, fruits, vegetables, legumes, nuts, and olive oil, may contribute to reduced chronic inflammation and potentially lower colorectal cancer risk [31].

Consistent with our finding, previous studies have reported a positive association between high uPDI scores and increased risk of CRC [14, 15]. A case-control study by Wu et al. involving 2799 CRC cases and 2799 age- and sex-matched controls further supports this observation, demonstrating that the odds of developing CRC increased with higher uPDI scores [37].

In contrast to hPDI, uPDI emphasizes high glycemic index (GI) foods like refined grains and sugar-sweetened beverages, which are typically lower in fiber [38]. This dietary pattern (uPDI) may contribute to CRC development through several mechanisms, including the stimulation of insulin secretion and hyperinsulinemia via the phosphatidylinositol 3kinase (PI3K)/ protein kinase B (AKT) signaling pathway [39]. Hyperinsulinemia, a condition of elevated blood insulin levels, can contribute to insulin resistance [39]. Insulin resistance can further promote CRC by stimulating growth-promoting effects in colon cells [40].

Our study found no significant association between overall adherence to the PDI and CRC risk. In contrast, higher adherence to the PDI was associated with a lower risk of CRC in Chinese women [37]. A meta-analysis study also reported no significant association between a vegetarian diet and CRC risk compared to a non-vegetarian diet [41]. Although no significant relationship was indicated between PDI and CRC risk, plant-based diets are good sources of certain nutrients, including folate, magnesium, and B vitamins [42, 43]. Folate plays a fundamental role in cellular deoxyribonucleic acid (DNA) methylation and cellular epigenetic changes. This function of folate introduces it as an essential factor in the correction of cell divisions as well as an inhibitory factor against cellular disorders [44].

To our knowledge, this is the first study to investigate the association between plant-based dietary indices and CRC risk in a Middle Eastern population. We employed a reliable and valid FFQ to minimize measurement error. Also, using valid instruments and controlling for multiple potential confounders were other strengths of this study. However, selection bias is inevitable in case-control studies. To mitigate this, we recruited cases and controls from the same referring hospital. Additionally, recall bias is inherent to the case-control design.

Conclusions

Our study found that greater adherence to the hPDI was significantly linked to a reduced risk of CRC, while higher uPDI scores were associated with an increased risk. This suggests that the quality, not just the quantity, of plant-based foods plays a significant role in CRC risk. Future studies with larger, diverse populations and prospective designs are warranted to confirm these findings and elucidate the underlying mechanisms by which hPDI and uPDI influence colorectal cancer risk.