Cancer is a leading cause of death in both more and less economically developed countries. The burden of all types of cancer is expected to grow worldwide due to population growth and aging [1]. Approximately 18.1 million new cancer cases and 9.6 million deaths occurred in 2018 worldwide [2]. Breast cancer (BC) is the most common type of cancer; it affects women throughout the world, and is among the top 5 cancers in Jordan [3]. In 2014, there were 1,187 newly diagnosed cases of BC in both genders, accounting for 20.8% of all newly diagnosed cancer cases [3].

Although a number of risk factors have been identified for BC, some are difficult to modify [4]. While dietary and nutrient intake patterns, physical activity (PA), alcohol drinking, body mass index (BMI), and smoking are modifiable risk factors of BC, genetic factors, age, and gender are considered to be non-modifiable risk factors [1,4,5]. There is accumulating evidence associating combined nutrient and dietary patterns, single micronutrients, macronutrients, and food items with the pathogenesis of BC [4,6,7]. However, most of these relationships remain controversial [4,6,7]. Although limited evidence is currently available in support of probable causal judgments regarding the etiological role of specific dietary factors [8], diet nonetheless plays a role in the continuous increase of BC incidence [9]. The associations of dietary and nutrient intake patterns with BC have been investigated in numerous studies [1,9-11]. The findings of these studies were inconclusive, and most studies have examined the risks associated with specific foods and nutrients, rather than measures of overall dietary patterns [1, 9-11]. Therefore, the objective of this study was to investigate associations between nutrient intake patterns and BC risk among Jordanian women.

The study was conducted in accordance with the ethical standards of the responsible committee on human experimentation and with the Helsinki Declaration of 1975 and subsequent amendments. The proposal was approved by the Institutional Review Board of the 2 participating hospitals: King Hussein Cancer Center (KHCC) and Al-Basheer (16 KHCC 57 and 2,250, respectively).

The results of this study add to the evidence for an association between BC risk and diet and lifestyle behaviors. In our study, a significant difference between cases and controls in BMI was detected; furthermore, the risk of BC increased as the BMI increased from normal body weight to overweight or obesity. BMI has been found to be associated with the risk of BC incidence, and this association has been reported in many studies [8,15-17]. Additionally, the present study revealed an inverse association between PA and BC. This is in agreement with several studies indicating that PA was significantly associated with a reduced risk of BC [18-21]. PA might reduce BC risk through several biological mechanisms, including its impact on adiposity, modulation of sex hormones, and reduction in insulin resistance, adipokines, and inflammatory markers [18-21].

Regarding diet, 3 nutrient intake patterns were identified in this study: a high vitamin C and β-carotene nutrient pattern, which was associated with an increased risk of BC; a high calcium, phosphorus, and vitamin D nutrient pattern, which showed a protective effect against BC; and a high-fat nutrient pattern, which was also associated with a higher risk of BC.

Although several studies demonstrated that vitamin C and β-carotene may favorably influence BC risk because of their antioxidant properties [22-25], our study showed that high intake of those 2 nutrients may be associated with higher risk of developing BC among Jordanian women. However, there is somewhat weaker evidence for a link between vitamin C and β-carotene with hormone-sensitive BC [24]. A plausible explanation for the association between antioxidants and BC risk was suggested by Salganik [26]. Those authors reported that reactive oxygen species in moderate concentrations act as indispensable mediators of cancer-protective apoptosis and phagocytosis, and that in people with a low level of reactive oxygen species, an excess of antioxidants can block these cancer-preventive mechanisms and promote cancer [26]. An excess of antioxidants, which interferes with apoptosis, can be cancer-promoting in people who are regularly exposed to the effect of environmental carcinogenic factors (tobacco smoke, industrial pollutants) that result in a high accumulation of pre-cancerous and cancerous cells [26]. Nagel et al. [27] demonstrated no associations between BC risk and dietary vitamin C and β-carotene. In contrast, Bakker et al. [24] documented that higher concentrations of plasma vitamin C and β-carotene reduced the risk of BC.

The second nutrient pattern in our findings was characterized by high calcium, phosphorus, and vitamin D intake, which may reflect high levels of dairy product consumption. According to our results, this nutrient pattern was significantly protective against BC. This finding is in agreement with the previous findings of Knekt et al. [28], who demonstrated that milk consumption exerted a protective effect against the risk of BC. Furthermore, milk comprises a complex mixture of macronutrients and micronutrients. The observed association may be caused by the protective effect of some or a combination of these nutrients [28]. This is also similar to results of Dong et al. [29], who documented that increased consumption of total dairy food, but not milk, may be associated with a reduced risk of BC. Other studies found that dairy consumption was not strongly related to BC risk [30,31]. Zhang et al. [31] supported the hypothesis that a high intake of dietary calcium may exert a protective effect on BC risk.

In the third nutrient pattern, which was characterized by high fat intake, a higher intake of MUFAs and PUFAs was found to significantly increase the risk of BC. In particular, among MUFAs, the trans form is likely to have been responsible for this relationship. The association between BC risk and trans fatty acids in general and MUFAs in the trans form was precisely studied by Chajès et al. [32]. Those authors found that intake of the cis forms of MUFAs was unrelated to BC risk, while a high serum level of trans MUFAs, presumably reflecting a high intake of fried foods, likely contributed to an increased risk of invasive BC in women [32]. A finding in this study (not presented in the Results section) was that trans fat intake in cases was higher than in controls and the percentage of total trans fats among total nutrients intake showed a significant trend (p<0.05) for BC risk. Furthermore, Wirfält et al. [33] concluded that PUFA intake increased the risk of BC. PUFAs produce eicosanoids, which increase the risk of BC. Besides, PUFAs have many double bonds that are easily oxidized, enhancing cellular damage [33]. Similarly, Chajès et al. [32] documented that PUFA intake significantly increased the risk of BC. However, Sczaniecka et al. [34] found that total PUFA intake was not associated with BC. Farvid et al. [35] reported the same pattern in their study. They found that a higher intake of fats was associated with a higher risk of BC; in addition, intake of saturated fatty acids (SFAs) and MUFAs was also related to an increased risk of BC among all participants. This was in agreement with the results of Salarabadi et al. [15], who suggested that SFA intake increases the risk of BC. In contrast, the findings of Alexander et al. [36] were not supportive of a positive independent association between consumption of SFA and BC.

In a study of this type, we rely greatly on the ability and memory recall of participants to accurately and carefully provide information from a period when it was not necessarily important to remember the details of meals that were consumed, or PAs that were undertaken. It is understandable that some individuals may have a better recall than others, and that biases may exist in the minds of those being interviewed, and indeed, in the interviewer. We also did not take into account the possible effects of cooking on the bioavailability of the various nutrients, although we attempted to control for a range of potential confounders. Alcohol consumption was not measured because its consumption is culturally discouraged in Jordan. Only 2 women reported alcohol consumption. Although the sample size was small, it was representative, as confirmed by statistical calculations. The enrollment of the participants utilized convenience sampling, and was based on their agreement to participate in the study, which imparts some degree of bias in terms of sample selection. Jordan is a small country with an estimated population in 2012 of approximately 9.5 million people, and a cancer diagnosis rate for BC of 32.1 per 100,000. In 2014, a total of 1,187 newly diagnosed BC cases in both genders were reported by the Jordanian Cancer Registry. We had difficulties in recruiting more participants to increase our sample size.

A major strength of our study is the use of a validated and detailed FFQ to collect dietary data from our study population. Even though dietary data were collected only once, the FFQ has been reported to be an adequate instrument for measuring macronutrient and micronutrient intake. However, confirmative studies should verify and extend the presented data on Jordanian dietary habits in order to establish recommendations for people in Jordan to reduce BC incidence.

In conclusion, 3 nutrient intake patterns were identified in this study: a high vitamin C and β-carotene nutrient intake pattern, which was associated higher BC risk; a high calcium, phosphorus, and vitamin D nutrient intake pattern, which showed a protective effect against BC; and a high-fat nutrient intake pattern, which was also associated with a higher risk of BC.