Dietary habits, commensal microbiome, and nasopharyngeal carcinoma

Abstract

Nasopharyngeal carcinoma (NPC) is a malignant disease characterized by unique geographic distribution endemic to southern China, Southeast Asia, and the Middle East/North Africa. It has been widely accepted that the interaction of Epstein-Barr Virus infection, environmental and lifestyle factors, and genetic susceptibility, contributes to NPC carcinogenesis. In the past two decades, extensive application of intensity-modulated radiotherapy and widespread introduction of chemotherapy have significantly contributed to a desirable prognosis: better survival with fewer toxicities. However, there are still numerous knowledge gaps and unsolved questions about NPC risk and prognosis. In this thesis, we investigated whether dietary habits (Study I), and oral commensal microbiome (Study II) were associated with NPC risk, using a population-based case-control study in endemic southern China. We also conducted a longitudinal hospital-based NPC cohort study (Study III) to deliver proof-of-concept data on the commensal microbiome patterns in patients’ nasopharynx during radiotherapy and their role in NPC prognosis.

In Study I/Paper I, we analyzed a total of 4398 study participants (2174 NPC cases and 2224 controls) with data about adolescent diet and 4832 participants (2387 NPC cases and 2445 controls) with data about adulthood diet. We demonstrated a strong positive association between higher consumption of the “animal-foods-based diet” and NPC risk and a strong negative association with higher intake of the “plant-based diet”. Following mutual adjustment for adolescence and adulthood dietary patterns, risk estimates for the former were attenuated and no longer statistically significant, whereas associations with adulthood dietary patterns remained virtually unchanged.

In Study II/Paper II, we explored the relationship between NPC status and the oral microbiome using 16S rRNA sequencing in a study of 994 participants (499 NPC cases, 494 controls). We observed a significant reduction in community richness in NPC cases compared to that in controls. We also identified a pair of Granulicatella adiacens amplicon sequence variants (ASVs; Gran-7770 and Gran-5a37), which were strongly associated with NPC status and differed by a single nucleotide. We further revealed that Gran-7770 and Gran-5a37 each formed co-occurring nodes with a dozen ASVs, which were exclusive. These results suggest differences in the oral microbiomes between NPC patients and controls, which may be associated with both a loss of microbial diversity and niche specialization among closely related microorganisms.

In Study III/Paper III, we analyzed 445 nasopharyngeal samples longitudinally collected from 39 NPC patients during radiotherapy-based treatment. We addressed stable, temporal changes in the nasopharyngeal microbial community structure among NPC patients during treatment. These changes were associated with patients’ short-term clinical outcomes measured three months after the completion of radiotherapy. We also identified 23 out of 73 abundant ASVs that showed statistically significant changes in the ratio of abundance between early and late responders throughout treatment. These results provided evidence of an association between nasopharyngeal commensal microbiome and NPC patients’ short-term clinical outcome.

By studying a range of topics, this thesis provides more insights into NPC in terms of risk and prognosis: plant-based and animal-foods-based diets are differentially associated with NPC risk in endemic southern China, suggesting a possibility of primary prevention of NPC through dietary intervention; oral microbiome is associated with NPC risk; the niche specialization among closely related commensals associated with NPC status calls for future culture-based investigations; moreover, stable, temporal changes of the nasopharyngeal microbiome are associated with NPC patients’ short-term clinical outcome, which calls for more extensive longitudinal studies with long-term follow-up for verification and serves as a base of generating new hypotheses for future studies.