Introduction

Plasma cell disorders (PCDs) are a spectrum of diseases characterized by abnormal proliferation of plasma cells, from precursor conditions, monoclonal gammopathy of undetermined significance (MGUS) and smoldering multiple myeloma (SMM), to malignant multiple myeloma (MM). Over the past decade, therapeutic advances have improved survival and outcomes for MM patients. As individuals live longer, there are opportunities to enhance long-term health and quality of life through diet and lifestyle interventions.

Evidence suggests that dietary patterns and obesity are associated with risk and survival of PCDs [1]. High insulinemic and inflammatory diets, rich in processed foods, refined grains, and added sugar, are linked with increased risk of MM and MM-related mortality [2, 3]. Conversely, plant-based diets and higher intake of whole grains, fruits, and vegetables are associated with reduced risk of MGUS and MM [3,4,5]. Similarly, higher Healthy Eating Index (HEI) scores correlate with reduced all-cause and cancer-specific mortality [6, 7]. Given the limited data on nutrition in individuals with PCDs, our study examined dietary patterns in this population.

Methods

From May 2023 through January 2024, individuals with known histories of MGUS, SMM, or MM completed the self-administered 2014 full-length semiquantitative Block Food Frequency Questionnaire (FFQ) through the patient data portal of HealthTree Foundation, a 501©3 non-profit organization. Participants provided electronic informed consent before survey initiation. The FFQ evaluates the frequency of intake of 127 food and beverage items with portion sizes. Details about the questionnaire and its validation are available elsewhere (https://www.nutritionquest.com/validation-studies.html#full-length). De-identified FFQs were analyzed by NutritionQuest and statistically correlated with pre-collected demographic and health data retrieved from the HealthTree platform.

This study was reviewed by the Memorial Sloan Kettering Cancer Center Institutional Review Board and determined to be exempt from further review (IRB-A X22-043A(1)). All methods were performed in accordance with the relevant guidelines and regulations.

We excluded individuals with implausible dietary intake (n = 2), defined as values more than two interquartile ranges above the 75th percentile or below the 25th percentile of daily kilocalorie consumption. We calculated HEI-2020 scores, which include 13 food components and range from 0 to 100, with high scores indicating greater adherence to the 2020-2025 Dietary Guidelines for Americans (DGA) (Supplementary Table 1). We converted HEI components to standard dietary units for reporting (e.g., cup equivalents/1000 kcal). Using Kruskal–Wallis tests, we evaluated the differences in the distribution of scores across subgroups—age, sex, race, body mass index (BMI), PCD diagnosis, education, and marital status. P-values reflect comparisons excluding individuals with missing data. Additionally, using Z-test, we compared HEI-2020 scores from this study to HEI-2015 scores of the general U.S. population from the National Health and Nutrition Examination Survey (NHANES), across age and sex. HEI-2020 scoring is identical to HEI-2015, as the DGA did not change.

Results

Out of 443 participants, 71% were 60 years or older, 65% were female, 53% were diagnosed with MM, and 50% had a BMI >= 25 kg/m2. 66% were non-Hispanic White, 5% Hispanic, 4% non-Hispanic Black, and 3% non-Hispanic Asian. Race/ethnicity data were missing for 22%. The average calculated HEI score was 69 (Table 1).

Table 1 Healthy Eating Index (HEI) Scores Among Surveyed Plasma Cell Disorder Participants.
Full size table

There were significant differences in HEI scores by BMI, calculated by self-reported height and weight. The average HEI score was 72 for individuals underweight or normal BMI, 68 for those within the overweight category, and 65 for those within the obese category (p < 0.001). There was a trend showing an increase in average HEI scores by education level: 67 for individuals with less than a college education, 68 for college graduates, and 70 for those with a graduate or professional degree (p = 0.052). There were no significant differences in HEI scores by age, sex, race/ethnicity, PCD diagnosis, or marital status (Table 1).

We evaluated subgroup differences in HEI food group component scores most pertinent to PCD risk as supported by prior literature (Table 2). In our analysis, males reported consuming more calories (p < 0.001) and less fiber (p < 0.001) and vegetables (p < 0.001) than females. Hispanics and non-Hispanic Asian and Black respondents reported the lowest daily caloric intake (p = 0.019). Individuals reporting higher BMIs demonstrated lower fiber intake (p < 0.001) and trended towards lower vegetable (p = 0.054) and more added sugar (p = 0.061) intake. Participants with precursor conditions, as compared to MM diagnosis, reported lower fruit intake (p = 0.040) and trended toward higher vegetable (p = 0.084) and lower added sugar (p = 0.099) intake. Lastly, higher education levels were associated with higher fiber intake (p = 0.046) and trended towards higher total fruit (p = 0.089) and lower added sugar (p = 0.093). There were no significant associations across all subgroups for whole fruit, whole grain, and dairy intake (not shown).

Table 2 Average Daily Food and Micronutrient Intake Among Surveyed Plasma Cell Disorder Participants.
Full size table

The average HEI-2020 score for the general U.S. population (age 2 + ) is 58. Study participants had higher HEI scores than the U.S. population overall and across age and sex [8, 9].

Discussion

In relation to modifiable risk factors, individuals with higher BMI (obesity classification) reported intake consistent with a lower HEI score, reflecting lower diet quality. This same pattern was shown for fiber intake, and a trend was seen with lower vegetable and higher added sugar intake. Given the association between obesity and increased MM mortality as well as progression from MGUS to MM [10,11,12], these individuals may benefit from dietary interventions aimed at improving HEI scores.

A higher education level correlated with better adherence to nutritional guidelines, reflected by significantly higher fiber intake and a trend for higher HEI scores, total fruit consumption, and lower added sugar intake. The association between education level and a wide range of health outcomes underscores the importance of interventions that address health literacy and socioeconomic barriers to adopting healthier diets and adhering to dietary guidelines for at-risk populations (e.g., food insecurity, food apartheid/deserts).

We found notable dietary differences by sex, race/ethnicity, and type of PCD. Compared to females, males reported higher caloric intake and lower fiber and vegetable consumption. These findings align with prior studies showing sex-based differences in dietary habits [13], which may influence health outcomes. Although HEI scores were similar between participants with precursor conditions and MM, and both groups met recommended total fruit, total vegetable, and added sugar intake goals per the DGA, those with precursor conditions had significantly lower fruit intake, with a trend towards higher vegetable and lower added sugar intake. These modest differences may reflect variations in how patients interpret and prioritize dietary recommendations based on their disease stage and treatment status. Additionally, the higher fruit intake and trend toward higher added sugar in MM patients may be partly explained by glucocorticoid use, a mainstay of MM treatment, which may influence dietary preference for highly palatable foods high in sugar and fat [14], but the clinical significance of these patterns warrants further investigation. Racial/ethnic differences in caloric intake were also observed, potentially reflecting limitations of FFQs in capturing culturally specific dietary behaviors, patterns, and food items. However, interpretation is limited by the small number of non-White participants.

A limitation of this study is selection bias, which may explain the higher HEI score in our population compared to the general U.S. population. Recruitment largely occurred through nutrition-focused talks hosted by support groups and the HealthTree Nutrition and Wellness Chapter, likely attracting individuals with greater interest and awareness of nutrition. A recent study of diet quality among cancer survivors, including those with MM, found lower average diet quality scores than the general U.S. population [15]. However, our prior research suggests individuals with PCDs adopt more healthful diets post-diagnosis, likely in response to the morbidity and mortality risks associated with their condition [16]. Additionally, methodological variation may also explain differences, as that study used two 24-hour recalls, which have limited ability to capture less frequently consumed foods [17], whereas we chose the Block FFQ to assess usual intake over the past year, which is more relevant for understanding habitual eating patterns. Although previous research supports the FFQ’s reproducibility and accuracy compared to 24-hour recall [18], this methodological difference should be considered when interpreting results. Our study may also be influenced by reporting bias, a common limitation of dietary self-report [19, 20]. Participants with elevated BMI reported similar or lower caloric intake than those with normal BMI. However, the dietary components we present, normalized per 1000 kcal, are less sensitive to such bias and provide a more reliable measure of diet quality. Additionally, the FFQ introduces recall bias that may affect the accuracy of reported intake.

This is the largest study to date examining the dietary habits of individuals with PCDs, a population that may benefit from diet modifications. The FFQ’s self-administered format enabled us to recruit a large sample. Using HEI-2020 scores and standardized nutrient measures, our methodology provides a robust foundation for comparing scores across samples and identifying areas where dietary improvements are needed. These findings may support dietary counseling, survivorship care, and public health strategies for patients with PCDs. This also presents an opportunity for healthcare providers to encourage lifestyle modifications. Although providers recommend changes more often to patients with cancer than those without, only about 50% of patients overall receive health behavior guidance [21]. Consistent with this, our prior study found that 57% of patients with PCDs reported that diet and nutrition were not addressed during hematology or oncology visits [16]. These findings underscore the need for healthcare providers to consistently integrate dietary counseling into routine care, particularly as patients are motivated to improve their diet after diagnosis [16].

Conclusion

Dietary patterns varied across subgroups of individuals with PCDs, with fiber intake being the most consistent differentiator. Lower fiber intake was reported among males, higher BMI categories, and lower education levels. Elevated BMI was also associated with lower HEI scores. Together, our findings suggest that these subgroups may benefit from targeted dietary interventions. This study highlights the importance of collaborative care between medical providers, nurses, dietitians, and patient support groups to address dietary habits and health literacy in at-risk populations. However, further prospective research is needed to understand the role of dietary behaviors in biomarker changes, disease progression, survival, and quality of life outcomes in patients with MM and its precursor disorders (such as the five NUTRIVENTION trials) [22, 23].