Introduction
With the exception of some rare genetic conditions that increase the risk of cancer, the etiology of childhood and adolescent cancer is largely unknown. Reference Wigle, Arbuckle and Turner1 It has long been hypothesized that risk factors for childhood cancers may present during the prenatal and neonatal period. However, only very few exposures during these periods have been robustly associated with childhood or adolescent cancer, including the association of radiation exposure with leukemia and maternal use of diethylstilbestrol with vaginal and cervical cancer in female offspring. 2 Numerous other factors like maternal tobacco smoking, alcohol consumption, supplemental folic acid in pregnancy, occupational exposures, and industrial pollutants Reference Linet, Wacholder and Zahm3 are suspected to be associated with increased risk of childhood and adolescent cancer, but epidemiological studies are inconsistent.
Nasopharyngeal carcinoma (NPC) arises from the nasopharyngeal mucosal lining and is a rare malignancy in children and adolescents. Southern China has the highest incidence rates of childhood and adolescent NPC in the world. According to the Cancer Registry of Hong Kong, NPC is the fourth most common malignant neoplasm in men and the fifth most common malignant neoplasm in women during the second decade of life. 4 Childhood and adolescent NPC have distinct clinical characteristics. For example, compared to adults with NPC, children and adolescents with NPC are more likely to present with advanced stage disease but have a better prognosis. Reference Wei and Sham5,Reference Ayan, Kaytan and Ayan6 The overall survival of pediatric patients after combined chemo- and radiotherapy has been reported to exceed 75% in most studies. Reference Ozyar, Selek and Laskar7,Reference Lu, Chang and Sun8 Unlike adult NPC, childhood and adolescent NPCs are more likely to be associated with Epstein–Barr virus (EBV) infection. Reference Wei and Sham5,Reference Ayan, Kaytan and Ayan6 These differences highlight the need for a better understanding of childhood and adolescent NPC. To the best of our knowledge, no study to date has evaluated the association between exposures during the prenatal and neonatal period with the risk of childhood and adolescent NPC. The objective of this study was thus to analyze the association between exposures during the prenatal and neonatal period and the risk of childhood and adolescent NPC in a case-control study in southern China.
Method
Clinical data of patients with NPC referred to Sun Yat-sen University Cancer Center (SYSUCC) between January 2009 and January 2016 were reviewed. The SYSUCC classified disease according to the International Classification of Disease version 10 during this period, and NPC was identified as C11. All of the included patients met the following criteria: (1) pathologically confirmed World Health Organization (WHO) type I, II, or III NPC; (2) less than 18 years of age. The exclusion criteria were that patients with a history of synchronous or previous another type of malignancy. Controls selected from the Sun Yat-sen University Zhongshan Ophthalmic Center were the childhood, and adolescent (i.e., less than 18 years of age) patients had not a history of malignancy. The two groups were frequency-matched for age and gender.
The following information was ascertained from the biological mothers of cases and controls by trained interviewers using a structured questionnaire by telephone interviews between September 2019 to December 2019: parental smoking, maternal alcohol consumption, diseases, use of medications and/or supplements, or environmental exposure during pregnancy, and obstetric outcomes (gestational weeks, delivery mode, and birth weight). To reduce recall bias, the obstetric outcomes (such as gestational weeks and birth weight) collected through telephonic interview were verified from the certificate of birth and immunization, if parents could provide them. The following baseline characteristics were recorded from the medical charts: age, gender, family history of malignancy, Tumor Node Metastasis stage, levels of serum EBV antibodies [viral capsid antigen-immunoglobulin A (VCA-IgA) and early antigen-immunoglobulin A (EA-IgA)], and plasma EBV DNA.
Maternal smoking history before or during the pregnancy with the index child and adolescent was ascertained from self-reported data; if mothers smoked, their average daily consumption of cigarettes was recorded. Information regarding paternal consumption of cigarettes was also collected. Maternal alcohol intake during pregnancy was also assessed. Additionally, the respondents were asked if they had used medications or supplements (folic acid and/or multivitamin) or had a history of relevant diseases (anemia, gestational diabetes, and gestational hypertension) during pregnancy. To assess environmental exposure, we asked the mothers whether they had a history of exposure to radiation (e.g., radiographs), exposure to toxins (e.g., chemical fertilizers and pesticides), exposure to paint, and intimate contact with pets during pregnancy (daily contact with pets during pregnancy). We also assessed gestational age at delivery (full-term birth, 37–41 weeks; premature birth, 28–36 weeks; post-term birth, ≥42 weeks), delivery mode (vaginal or cesarean), and birth weight (To reduce recall bias, we focus on common birth weight classifications: low birth weight, <2500 g; normal birth weight, 2500–3999 g; high birth weight, ≥4000 g). Information regarding a family history of malignancy in the control group was also collected. Informed consent was provided by the parents of all participants in the childhood and adolescent NPC group and the control group.
Pearson’s chi-square test or Fisher’s exact test was used to assess the differences in variables between the childhood and adolescent NPC group and the control group, and the Mann−Whitney U test was used for the non-normally distributed continuous variable (age). The associations between variables and the risk for childhood and adolescent NPC were evaluated using unconditional logistic regression analysis, from which odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were derived. All the tests were two-sided, and 0.05 was set as the significance level. Data analysis was performed using SPSS 23.0 for Windows (Chicago, IL). The authenticity of this article has been validated by uploading the key raw data onto the Research Data Deposit public platform (www.researchdata.org.cn), with the approval RDD number RDDA2021540761.
Results
A total of 112 childhood and adolescent NPC at the Department of Nasopharyngeal Carcinoma of SYSUCC from January 2009 to January 2016 were screened. Among the 112 patients, 6 died, 15 were lost to follow-up, 17 refused to participate, and 28 were excluded because of communication barriers or failure to recall relevant information during pregnancy. Finally, 46 patients were enrolled for the analysis. Forty-five age- (±2 years) and sex-frequency-matched controls from the Sun Yat-sen University Zhongshan Ophthalmic Center agreed to join this study.
NPC patients’ characteristics
The major clinical characteristics of the patients with childhood and adolescent NPC are shown in Table 1. The median age was 14.5 years, with a male-to-female ratio of 2.07:1. All the patients were pathologically WHO type III NPC. At the time of diagnosis, 12 (26.1%) patients had stage III disease, 31 (67.4%) had stage IVa–b, and 3 (6.5%) had stage IVc disease with bone metastasis. Three (6.5%) patients had a family history of cancer in first-degree relatives and 10 (21.7%) in second- or third-degree relatives. One patient had a family history of NPC in second-degree relatives with no affected first-degree relatives. Information about baseline plasma EBV DNA and serum EBV antibodies was available in 40 and 41 patients, respectively. Overall, positive plasma EBV DNA was found in 32 (80%) patients, and more than half (24, 60%) had plasma EBV DNA ≥ 4000 copies/ml. Positive VCA-IgA and EA-IgA were detected in 36 (87.8%) and 20 (48.8%) patients, respectively. The highest serum VCA-IgA and EA-IgA titers observed were 1:1280 and 1:160, respectively.
NPC patient demographics and clinical characteristics
EA, early antigen; EBV, Epstein–Barr virus; IgA, immunoglobulin A; VCA, viral capsid antigen.
Maternal exposures during pregnancy and obstetric variables
The characteristics of maternal exposures during pregnancy and obstetric variables of the childhood and adolescent NPC group and the control group are described in Table 2. The median age of the patients with childhood and adolescent NPC and controls during the study period was 22 and 23, respectively. Of the 45 controls, 26.7% were female and 73.3% were male. Compared to controls, patients with childhood and adolescent NPC were more likely to have a family history of cancer (28.3% of cases and 11.1% of controls). None of the mothers included in the study reported smoking before or during pregnancy, and only one had consumed alcohol during pregnancy. Forty-one fathers (25 from the childhood and adolescent NPC group, 16 from the control group) smoked before and during pregnancy, with the majority (61.0%) smoking more than half a packet of cigarettes per day.
Maternal exposure factors during pregnancy and obstetrical indicators
* Fisher’s exact test.
# Chi-square test continuity correction.
^Children and adolescents.
Maternal use of medications during pregnancy was reported more frequently by the mothers of patients with childhood and adolescent NPC (17.4%) than controls (4.4%). Anticold medications were the most commonly used, followed by traditional Chinese medicine. There was a low rate of maternal use of folic acid and/or multivitamins (6.5%) during pregnancy among the mothers of patients with childhood and adolescent NPC. Only one mother appeared anemia in the gestation period. No mothers in either group had a history of exposure to radiation during pregnancy, and 1 and 2 mothers of patients with childhood and adolescent NPC and controls, respectively, had intimate contact with pets during pregnancy. Compared with the controls, the mothers of patients with childhood and adolescent NPC were more likely to have been exposed to toxins or paint (4.3% of cases and 0% of controls) during pregnancy.
The overall cesarean section rate of mothers of cases and controls was 19.6% and 20%, respectively. Almost all the patients with childhood and adolescent NPC were born at full term (91.3%), with 1 (2.2%) born premature and 3 (6.5%) born post-term. In the control group, 42 (93.3%) were born at full term, 3 (6.7%) were born premature, and there were no post-term births. The incidence of normal birth weight was similar between the childhood and adolescent NPC group (95.7%) and the control group (93.3%).
The association of maternal exposures during pregnancy and obstetric variables with childhood and adolescent NPC risk
Tables 3 and 4 show the association analysis between maternal exposures and obstetric variables with the risk of childhood and adolescent NPC. In univariate analysis, the maternal use of folic acid and/or multivitamins during pregnancy was associated with a reduced risk of childhood and adolescent NPC in the offspring (OR = 0.07, 95% CI = 0.02–0.25, P < 0.001). A family history of cancer was associated with a higher risk of childhood and adolescent NPC compared to no family history of cancer (OR = 3.15, 95% CI = 1.02–9.75, P = 0.046). We also found an increasing albeit nonstatistically significant trend toward an increased risk for childhood and adolescent NPC with maternal use of medications during pregnancy (OR = 4.53, 95% CI = 0.91–22.64, P = 0.066) and paternal smoking before or during pregnancy (OR = 2.16, 95% CI = 0.93–5.01, P = 0.073). No association was observed with intimate contact with pets during pregnancy (OR = 0.48, 95% CI = 0.04–5.46, P = 0.552), delivery mode (OR = 0.97, 95% CI = 0.35–2.73, P = 0.958), and birth weight (OR = 0.64, 95% CI = 0.10–4.00, P = 0.630). After multivariable analysis, only the trend of the maternal use of folic acid and/or multivitamins during pregnancy for reduced risk of childhood and adolescent NPC remained significant (OR = 0.07, 95% CI = 0.02–0.26, P < 0.001; Table 4).
Univariate analysis of maternal exposure factors and obstetrical indicators with childhood and adolescent nasopharyngeal carcinoma
Multivariable analysis of maternal exposure factors and obstetrical indicators with childhood and adolescent nasopharyngeal carcinoma
* Adjust for family history of malignancy and mother supplemental folic acid and/or multivitamin during pregnancy.
Discussion
To our knowledge, this is the first study to evaluate the association between maternal exposures during pregnancy and obstetric variables with the risk of childhood and adolescent NPC. We found that the maternal use of folic acid and/or multivitamins during pregnancy was associated with a reduced risk of childhood and adolescent NPC in the offspring.
Numerous studies have suggested an inverse association between the intake of folic acid and/or multivitamins and the risk for adult cancer. Reference Gaziano, Sesso and Christen9 A previous large-scale, double-blind, randomized, placebo-controlled trial by Gaziano et al. showed that compared to placebo, men taking a daily multivitamin had a modest but significant reduction in the risk of total cancer. Reference Gaziano, Sesso and Christen9 Numerous studies have indicated that maternal and prenatal use of vitamins and/or folic acid reduces the risk of various pediatric cancers, including acute myeloid leukemia, acute lymphoblastic leukemia, neuroblastoma, and brain tumors. Reference Metayer, Milne and Dockerty10–Reference Goh, Bollano, Einarson and Koren12 Nevertheless, few studies have assessed the association between the use of vitamins and/or folic acid and the risk for NPC, particularly for childhood and adolescent NPC. A matched case-control study from Zeng et al. suggested that dietary folate and vitamin B6 may be protective for adult NPC in southern China. Reference Zeng, Liu and Lin13 Similarly, our study showed that the maternal use of folic acid and/or multivitamins during pregnancy was inversely related to the risk of childhood and adolescent NPC in the offspring.
There are several possible mechanisms underlying the relationship between the maternal use of folic acid and/or multivitamins during pregnancy and childhood and adolescent NPC risk. One mechanism supported by many studies was that folic acid and/or multivitamins influence genetic and epigenetic processes involved in fetal development through one-carbon metabolism. Reference Christman, Sheikhnejad, Dizik, Abileah and Wainfan14–Reference Steegers-Theunissen, Obermann-Borst and Kremer17 One-carbon metabolism, a metabolic network centered around the folate and methionine cycles, is vital for the mutual conversion of serine and glycine, de novo thymine synthesis, de novo purine synthesis, remethylation of homocysteine to methionine, and the generation of s-adenosylmethionine (SAM), a methyl donor used widely for biological methylation reactions. Reference Tibbetts and Appling18–Reference Shane20 As cancer cells are highly proliferative cells with an increased nucleotide requirement, the one-carbon metabolism is critical for cancer cells. Folic acid, a water-soluble B vitamin, is derived from food and converted into tetrahydrofolate in the body. Recent studies have suggested that folate-mediated one-carbon metabolism impacts epigenetic and redox homeostasis processes in cancer cells. Reference Ducker and Rabinowitz21,Reference Davis and Uthus22 Folate deficiency may lead to decreased tissue SAM, global DNA hypomethylation, and ultimately hepatic tumorigenesis in rodents in the absence of carcinogen treatment. Reference Davis and Uthus22 Furthermore, a study by Liu et al. found that folic acid inhibits NPC cell proliferation and invasion via activation of the FRα/ERK1/2/TSLC1 signaling pathway. Reference Liu, Jin, Pi and Liu23 Vitamin B12 and B6 also play an important role as coenzymes in the one-carbon metabolism. Reference Christman, Sheikhnejad, Dizik, Abileah and Wainfan14 Another possible mechanism is that maternal nutritional status before and during pregnancy may play an important role in the establishment of the child’s immune system. Reference Ramakrishnan, Grant, Goldenberg, Zongrone and Martorell24,Reference Palmer25 Maternal nutritional imbalances, both deficiency and excess, may alter the immune system of the offspring and may result in immune-mediated diseases such as chronic inflammatory conditions like cancer. Reference Ramakrishnan, Grant, Goldenberg, Zongrone and Martorell24,Reference Palmer25
In the current study, we also found an increasing albeit nonstatistically significant trend toward an increased risk for childhood and adolescent NPC with maternal use of medications during pregnancy, paternal smoking before or during pregnancy, and a family history of cancer. It has been observed that maternal use of medications that alter amino acid conjugation (predominantly antipyretics and analgesics) during pregnancy significantly increased the relative risk for childhood cancer. Reference Gilman, Wilson, Kneale and Waterhouse26 Paternal, but not maternal, smoking around the time of conception has also been reported as a risk factor for childhood cancer, which may be explained by the lower risk of transmission of embryonic genetic mutations from the mother than from the father. Reference Milne, Greenop and Scott27,Reference Guerquin, Duquenne and Coffigny28 Our findings are also in line with those of previous studies which have reported a strong association between the risk of adult NPC and a family history of NPC and other cancers, such as head and neck cancer, lung cancer, and breast cancer. Reference Ren, Liu and Qin29,Reference Jia, Feng and Xu30
This study has several limitations. First, this was a retrospective study and recall bias may affect our findings. Second, its single institutional design, small sample, sampling method, matching method, and lack of information about the mothers’ postnatal lifestyle may have caused a significant bias and led to statistical uncertainty in the estimated associations. Therefore, further independent validation of our findings is warranted.
In conclusion, this is the first investigation to evaluate the association between maternal exposures during pregnancy and obstetric variables with the risk of childhood and adolescent NPC. Our results show that the maternal use of folic acid and/or multivitamins during pregnancy was associated with decreased risk of childhood and adolescent NPC in the offspring.
Acknowledgements
This work was supported by grants from the Fundamental Research Funds for the Central Universities, Sun Yat-sen University (20ykpy174) and Natural Science Foundation of Guangdong Province (2020A1515110472).
Financial Support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Conflicts of Interest
None.
Ethical Standards
The study was approved by the Institute Review Board of Sun Yat-sen University Cancer Center (GZR2017-100).
