Consumption of processed food and risk of nasopharyngeal carcinoma: a systematic review and meta-analysis.

Background: Consumption of processed foods has been associated with nasopharyngeal carcinoma (NPC), but with inconsistent results. Therefore, we conducted a systematic review and meta-analysis to compute results regarding the association between processed foods and risk of NPC in included studies.
Methods: Studies exploring the association between consumption of processed food and risk of NPC were included in the present study. All included studies were case-control or cohort designed. PubMed, Web of Science, EMBASE, Medline and Google Scholar databases were searched for articles published before July 2021. We recorded the following data: author, publication year, sample size, study type, study location, years of diagnosis, food item and comparison, and the covariates considered were multivariate adjusted odds ratios (ORs) or relative risks (RRs) with corresponding 95% confidence intervals (CIs) for the highest vs. lowest categories of processed food intake. STATA 12.0 software was used to compute the multivariate ORs or RRs and 95% CIs of the association. Quality appraisal was made using the Cochrane Risk of Bias Tool.
Results: A meta-analysis was made for 29 case-control studies (including 14,378 NPC patients and 17,928 controls). The meta-analysis showed that the highest categories of processed food intake were associated with a 65% increase in NPC risk compared with the lowest categories in a random effects model (OR =1.67; 95% CI: 1.56-1.79; P value for Q test <0.001; I2=86.9%). Subgroup study showed significant positive associations regarding consumption of processed food and risk of NPC in both Asians and Caucasians (Asian: OR =1.68, 95% CI: 1.56-1.81; Caucasian: OR =1.36, 95% CI: 1.09-1.71). Conclusions: The association of processed foods with NPC risk might be significant. Further prospective studies and experimental research are needed to explore this relationship. 2022 Translational Cancer Research. All rights reserved.

Introduction

Nasopharyngeal carcinoma (NPC) is a malignancy with distinct geographical distribution, with an estimated 129,079 new cases and 72,987 deaths in 2018 worldwide, and most NPC patients are geographically localized to southern China and east and southeast Asia (1). According to relevant epidemiological data, the incidence and mortality of NPC in China were 3.09/100,000 and 1.57/100,000, respectively, which was higher the global data, and the top three incidence and mortality provinces are all located in southern China (2). NPC is a head and neck cancer with poor prognosis, and epidemiological trends in the past decades have shown the incidence and mortality of NPC have been decline (3). It is thought that NPC is the result of both genetic susceptibility and exposure to environmental factors such as Epstein-Barr virus infection (4).

Salt-preserved and fermented foods such as salted fish are traditional southern Chinese food, and favored by the local population due to the unique flavor of preserved foods and local food culture. Previous studies have proved that the intake of traditional salt-processed food is associated with gastrointestinal tumors including esophageal and gastric cancer (5,6). Pro-cancer factors in processed meat, including excess fat, excess protein, excess iron and heat-induced mutagens, may also be involved in carcinogenesis, plus the salt and nitrite added during the curing process (7). Several studies (8-20) have demonstrated that the consumption of processed foods is associated with risk of NPC. However, some inconsistent results exist. Some studies (21-24) showed no significant association between consumption of processed foods and risk of NPC. Therefore, we conducted a systematic review and meta-analysis to summarize the association of processed foods and NPC and assess the relationship. We present the following article in accordance with the MOOSE reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-22-690/rc).

Methods

Search strategy

PubMed, Web of Science, EMBASE, Medline and Google Scholar databases were searched by investigators for articles published before July 2021 regarding the association between consumption of processed food and risk of NPC. We used the search terms: (‘processed’ OR ‘pickle’ OR ‘pickled’ OR ‘moldy’ OR ‘fermented’ OR ‘saltextractible’ OR ‘salted’) AND (‘nasopharyngeal’) AND (‘cancer’ OR ‘carcinoma’ OR ‘neoplasm’ OR ‘neoplasia’ OR ‘neoplastic’). All articles were published in English.

Exclusion criteria

Studies exploring the association between consumption of processed food and risk of NPC were included in the present study. All included studies were case-control or cohort designed. Exclusion criteria were: (I) not provide information regarding the relative risk (RR) or odds ratio (OR) estimates and the 95% confidence interval (CI) for the association between consumption of processed food and risk of NPC; (II) not case-control or cohort design; and (III) reviews, meta-analyses and case studies were removed.

Data collection

We recorded the following data: author, publication year, sample size, study type, study location, years of diagnosis, food item and comparison, and the covariates considered were multivariate adjusted ORs or RRs with corresponding 95% CI for the highest vs. lowest categories of processed food intake.

Statistical analysis

STATA 12.0 software was used to compute the multivariate ORs or RRs and 95% CIs regarding the association between consumption of processed food and risk of NPC. Q test and I2 were used to evaluate heterogeneity between studies. With invariably high heterogeneity (P value for Q test ≤0.05 and I2≥50%), a random effects model was used to generate summary effect size of studies; conversely, in the absence of between-study heterogeneity (P value for Q test >0.05 and I2<50%), a fixed effects model was used to summarize the effect size. Meta-regression analysis was applied to explore source of heterogeneity. Subgroup analysis for different ethnicities was used to explore the effect of heterogeneous ethnicities on the heterogeneity of the meta-analysis. Subgroup analysis for studies adjusted for confounders and studies which did not report covariates was used to explore the source of heterogeneity. Sensitivity analysis was conducted by removing 1 individual study each time to assess the source of heterogeneity. Publication bias was evaluated with Egger’s regression test, Begg’s adjusted rank correlation test, and funnel plots. Quality appraisal was made using the Cochrane Risk of Bias Tool. Data were analyzed using Review Manager 5.3.

Results

Search results

shows procedure for exclusion and Table S1 shows the studies’ characteristics and results. A meta-analysis was made for 29 case-control studies (including 14,378 NPC patients and 17,928 controls). N=13 case-control studies (8-20) reported consumption of processed foods is associated with risk of NPC, whereas n=16 studies (21-36) showed no significant association between consumption of processed foods and risk of NPC.

Figure 1

Flow of information through the different phases of meta-analysis. NPC, nasopharyngeal carcinoma; RR, relative risk; OR, odds ratio; CI, confidence interval.

Meta-analysis results

The meta-analysis showed that the highest categories of processed food intake were associated with a 67% increase in NPC risk compared with the lowest categories in a random effects model (OR =1.67; 95% CI: 1.56–1.79; P value for Q test <0.001; I2=86.9%; ). Subgroup study showed significant positive associations regarding consumption of processed food and risk of NPC in both Asians and Caucasians (Asian: OR =1.68, 95% CI: 1.56–1.81; Caucasian: OR =1.36, 95% CI: 1.09–1.71; ). Meta-regression analysis indicated that publication year, gender and age were not responsible for heterogeneity across studies (publication year: P value =0.242; gender: P value =0.509; age: P value =0.837). Subgroup study showed significant positive associations between consumption of processed food and risk of NPC in studies adjusted for confounders and studies which did not report covariates (studies adjusted for confounders: OR =1.64, 95% CI: 1.52–1.76; studies which did not report covariates: OR =2.10, 95% CI: 1.65–2.67; ). A sensitivity analysis showed no changes in the direction of effect when any 1 study was excluded (). In addition, Begg’s test, Egger’s test and funnel plot showed no significant publication bias in the included studies (Egger’s test: P=0.066; Begg’s test: P=0.082; ). Risk of bias graph was showed in Figure S1. Details of the risk of bias summary was showed in Figure S2.

Figure 2

Forest plot of the association between processed food intake and NPC risk. OR, odds ratio; CI, confidence interval; NPC, nasopharyngeal carcinoma.

Figure 3

Subgroup study of the associations between processed food intake and NPC risk in Asians, Africans, and Caucasians. OR, odds ratio; CI, confidence interval; NPC, nasopharyngeal carcinoma.

Figure 4

Subgroup study of the associations between processed food intake and NPC risk in studies adjusted for confounders and studies which did not report covariates. OR, odds ratio; CI, confidence interval; NR, not reported; NPC, nasopharyngeal carcinoma.

Figure 5

Sensitivity analysis of the association between processed food intake and NPC. OR, odds ratio; NPC, nasopharyngeal carcinoma.

Figure 6

Funnel plot of the association between processed food intake and NPC. OR, odds ratio; CI, confidence interval; NPC, nasopharyngeal carcinoma.

Discussion

In this meta-analysis, we included 29 case-control studies with 14,378 NPC patients and 17,928 controls. Our results suggested a significant association of processed foods to NPC risk with a random effects model score showing OR of 1.67 at 95% CI of 1.56–1.79 (P<0.01). And for both Asians and Caucasians, processed foods were a high-risk factor for NPC.

Our result was consistent with that of previous several studies. Okekpa et al. reported that salted fish consumption was significantly associated with an increased risk in NPC (OR =1.41; 95% CI: 1.13–1.75) (37). Similarly, Li et al. reported a significant association between total processed meat consumption dose and risk of NPC, and the risk of NPC increased with increased consumption dose of processed meat (low-rank intake: RR =1.46, 95% CI: 1.34–1.64; moderate-rank intake: RR =1.59, 95% CI: 1.30–1.90; high-rank intake: RR =2.11, 95% CI: 1.31–3.42) (38).

Due to the process of salted preservation, salted fish can accumulate high levels of nitrosamines, which have a carcinogenic effect on multiple organs (39). In addition, extracts of processed food have reactivated Epstein-Barr virus in cell lines (40). Lau et al. showed that decreasing incidence and mortality of NPC in Hong Kong correlated with declining salted fish consumption (41). Barrett et al. also reported an increased risk of NPC with a high level of intake of hard Chinese-style salted fish during adolescence (OR =1.19, 95% CI: 1.03–1.39) (26). As well as salted fish, salted and picked vegetables have also been proved to be associated with elevated risk of cancer incidence, including gastric cancer, and NPC (42,43). These findings indicated a need to reduce the consumption of processed foods.

However, several study limitations should be acknowledged. First, although observation studies can suggest an association between processed foods and NPC, the molecular mechanisms of the pathogenesis of NPC caused by processed foods are still unclear. Second, in our study we did not discuss the effect of some variates, including the intake dose and duration of consumption of processed foods, whether from adolescence or as an adult. Third, the status of Epstein-Barr virus may affect the result regarding the association between processed foods and NPC, so the lack of Epstein-Barr virus affection may lead to an inaccurate conclusion.

Conclusions

In summary, our findings of this present study voted that the result for the association of processed foods with NPC risk might be significant. And further prospective studies and experimental researches are needed to explore the relationship between processed foods and NPC risk.