Association between red blood cell distribution width-to-albumin ratio and cancer prevalence in US adults: a NHANES cross-sectional study (1999–2018).
Qian
Wang
1✉
Emaillaurel_827@163.com
1
Department of Radiation Oncology
Liaocheng People’s Hospital
252000
Liaocheng
Shandong
China
Qian Wang1, *
Department of Radiation Oncology, Liaocheng People's Hospital, Liaocheng, Shandong, China, 252000.
E-mail: laurel_827@163.com
A
Abstract
Introduction:
The red blood cell distribution width-to-albumin ratio (RAR) is a composite biomarker reflecting integrated inflammation and nutrition status. Unlike conventional biomarkers assessing isolated pathways, RAR holistically evaluates their interplay, suggesting potential utility for cancer risk stratification. This study seeks to investigate the association between RAR and cancer in U.S. adults.
Methods
Using data from 43,094 NHANES participants (1999–2018), we employed multivariable logistic regression to assess RAR-cancer associations. Restricted cubic spline (RCS) evaluated nonlinear association and threshold effects. The study also used subgroup analysis and interaction tests to explore whether the association was stable in the population.
Results
In the cross-sectional study, 3,893 participants (9.0%) had cancer. RAR was positively associated with cancer among 43,094 participants aged ≥ 20 years. In the fully adjusted model, each per unit increase in RAR was associated with a 30% increase in the likelihood of cancer (OR = 1.30, 95% CI: 1.20 ~ 1.39, P < 0.001). Participants in the top quartile of RAR had a 36% increased risk of cancer than those in the bottom quartile of RAR (OR = 1.36, 95% CI = 1.22 ~ 1.53, P < 0.001). RCS revealed that the association between RAR and cancer was nonlinear (P for nonlinear = 0.028). Subgroup analyses showed that the association between RAR and cancer was significantly stronger in males group (P for interaction < 0.001).
Discussion
A
This study demonstrates that a significant correlation was identified between RAR and risk of cancer in U.S. adults, suggesting that RAR may function as a clinically relevant biomarker for risk stratification and provides potential evidence for subsequent pathological mechanism research. Further large-scale prospective studies are warranted to delineate the role of RAR in cancer.Keywords:
Nutrition
Inflammation
Red blood cell distribution width
Albumin
Cancer
NHANES
Cross-sectional study
A
A
A
A
1. Introduction
Cancer constitutes a leading global cause of mortality. According to the World Health Organization, there were over 19 million new cancer cases and nearly 10 million cancer-related deaths around worldwide in 2022, with incidence rates continuing to rise[1]。In recent years, the roles of inflammation and nutritional status in the development of cancer have garnered increasing attention. Numerous inflammation and nutrition - related indicators, such as the NPAR (neutrophil percentage to albumin ratio)[2, 3], HALP (hemoglobin, albumin, lymphocyte, and platelet score) [4–6], CALLY (C-reactive-protein-albumin-lymphocyte index)[7, 8], PNI (prognostic nutritional index)[9–12], and CAR (C-reactive protein to albumin ratio)[13–15], have been proven to have predictive value for the prognosis of malignant tumors. Moreover, most existing studies have focused on the utility of biomarkers in disease prognosis, while their ability to predict the early stages of cancer development has not been fully explored. Chronic inflammation can promote tumor occurrence and progression by altering the tumor microenvironment, promoting angiogenesis, and immune evasion[16, 17]. On the other hand, malnutrition or cachexia is associated with reduced survival rates in cancer patients[18]. Therefore, identifying biomarkers that reflect inflammatory and nutritional status is of great significance for cancer risk stratification and early intervention.
Red blood cell distribution width (RDW), a common parameter in routine blood tests, reflects red blood cell volume heterogeneity. Recent studies have found that an elevated RDW is not only an indicator of anemia but is also associated with systemic inflammation, oxidative stress, and the prognosis of various chronic diseases, such as cardiovascular diseases[19], diabetes[20], and malignant tumors[21]. Albumin, as a key plasma protein, reflects nutritional and systemic inflammatory status, and low albumin levels are significantly associated with shortened survival in cancer patients[22]. Based on this, the red blood cell distribution width - to - albumin ratio (RAR) may integrate dual signals of inflammation and nutritional imbalance, potentially serving as a novel predictor of cancer risk. Numerous studies have demonstrated that RAR has been widely associated with the development and progression of various diseases[23, 24, 25].Nevertheless, epidemiological studies on RAR and cancer prevalence remain scarce. Hence, we hypothesize that RAR may hold significant predictive value in identifying individuals at high risk of early - stage cancer.
In this study, we conducted a retrospective cohort study involving 43,094 adult participants from the National Health and Nutrition Examination Survey (NHANES) spanning the years 1999 to 2018. We used various analysis methods to assess the relationship between RAR and the risk of cancer. The secondary objective of this study is to investigate whether the relationship between these biomarkers and cancer prevalence follows a non-linear pattern and to identify threshold levels of the biomarkers that may influence cancer outcomes. The findings are expected to provide a novel biomarker for cancer screening and epidemiological evidence for the role of inflammation - nutrition interactions in cancer development.
2. Methods and materials
2.1. Study design
Data for this cross-sectional study were derived from the NHANES, a nationally representative research program conducted to systematically evaluate the health and nutritional status of adults and children in the United States. The study used data from ten NHANES cycles from 1999 to 2018, integrating demographic information, laboratory assessments, and questionnaire responses, ultimately yielding 43,094 participants met the study inclusion criteria.Participants were included if they had complete data on RDW, serum albumin levels, and cancer-related questionnaire responses.
A
A
The NHANES protocol received ethical approval through the Institutional Review Board (IRB) of the National Center for Health Statistics (NCHS), with all adult participants submitting written informed consent prior to participation. A
A
Our secondary analysis adhered to STROBE guidelines applicable to cross-sectional studies and were granted exemption from subsequent IRB review.2.2. Definition of exposure variable and outcomes
In our study, the independent variable was RAR. The serum albumin concentration (g/dL) was determined using the Bromocresol Purple method. The RDW (percentage) was measured by a Coulter analyzer in the mobile examination centers using peripheral blood. RAR was calculated by dividing RDW by the serum albumin concentration (RAR = RDW/albumin).
The diagnosis of cancer was determined based on responses to the question from the Medical Condition Questionnaire (MCQ): "Have you ever been told by a doctor or other health professional that you had cancer or a malignancy of any kind?" Trained interviewers posed the questions at home using the Computer-Assisted Personal Interviewing (CAPI) system, which was programmed with built-in consistency checks to reduce data entry errors.
NHANES includes a section on medical conditions that collects self-reported health information. We identified participants with a history of cancer or malignancy based on their responses to the question, "Have you ever been told by a doctor or other health professional that you had cancer or a malignancy of any kind?"
2.3. Definition of covariates
Based on the existing literature, various potential covariates considered in this study included age, gender, marital status, race/ethnicity, educational level, family income ( based on poverty-income ratio, PIR), body mass index (BMI), smoking status, drinking status, diabetes, coronary heart disease (CHD), and hypertension. Race/ethnicity included classifications such as non Hispanic white, non-Hispanic black, Mexican American, Other Hispanic, and other races. Marital status was categorized as married or living with a partner, widowed, divorced or separated,and never married. Educational attainment was classified as below high school, high school, and above high school. BMI was used to divide participants into underweight or normal weight (BMI < 25.0), overweight (BMI, 25.0–30.0), and obese (BMI >30.0) groups. Referring to previous studies, those who had smoked more than 100 cigarettes and were currently smoking were classified as current smokers, while those who had smoked more than 100 cigarettes but had quit were classified as former smokers. Drinking status was categorized as non-drinker or current drinker (≥ 12 standard drinks annually), with missing data coded as 'NA' and incorporated as a distinct categorical variable in analyses. The presence of previous diseases such as diabetes, coronary heart disease, and hypertension was based on questionnaire inquiries about whether participants had been informed by their doctor of these conditions in the past. The detailed definitions of the covariates are available at https://www.cdc.gov/nchs/nhanes/.
2.4. Statistical analysis
First, continuous variables are expressed as mean ± standard deviation (SD), while categorical variables are reported as frequencies and percentages (%). To evaluate differences among the various groups, we utilized the chi-square test or Fisher's exact test for categorical variables, the One-Way ANOVA test for normally distributed continuous variables.
Second, we used a logistic regression to estimate the association between the RAR and cancer. The RAR was specified as a categorical variable (Q1: 2.08–2.83; Q2: 2.83–3.05; Q3: 3.05–3.35; Q4: 3.35–10.22). We constructed 3 models: Model 1 was the crude model with no covariates adjusted. Model 2 was adjusted for gender, age, race and ethnicity, marital status, education, household income, and Model 3 was based on Model 2 and BMI, smoking status, drinking status, diabetes,coronary heart disease, and hypertension. The results of the multivariable logistic regression analyses were presented as both crude and adjusted odds ratio (OR) estimates with 95% Cis.
Third, subgroups were created according to age group, sex, BMI, race/ethnicity, educational level, smoking status, and drinking status. A stratified analysis was conducted to examine the association between RAR levels and cancer prevalence in model 3. The P values for interactions were evaluated through likelihood ratio tests by comparing logistic regression models with or without the cross-product terms for each assessed factor.
Fourth, we conducted restricted cubic spline (RCS) regressions to estimate possible nonlinearity between RAR as a continuous variable and cancer after adjusting for confounding factors (model 3).
Finally, In all analyses, statistical significance was defined as a two-sided p-value < 0.05. All analyses were conducted using Free Statistics software (version 2.1.1; Beijing Free Clinical Medical Technology Co., Ltd. https://www.clinicalscientists.cn/freestatistics/)
3. Results
3.1. Characteristics of Study Population from NHANES
A
As shown in Fig. 1 for inclusion and exclusion criteria, 43,094 qualified participants from NHANES 1999–2018 were included in this retrospective study. Of these, 39201 participants were without cancer and 3893 participants were cancer patients. The overall cancer prevalence was 9.0%. Table 1 shows the general characteristics of the participants according to the red blood cell distribution width-to-albumin ratio. The four groups differed in age, gender, race, marital status,education level, family income, BMI, smoking status, drinking status, diabetes,coronary heart disease, and hypertension (all p-values < 0.05).Fig. 1
The flow chart of the study.
|
Characteristic
|
Total (n = 43094)
|
Q1
(n = 10698)
|
Q2
(n = 10223)
|
Q3
(n = 11276)
|
Q4
(n = 10897)
|
p
|
|---|---|---|---|---|---|---|
|
Gender, n (%)
|
< 0.001
|
|||||
|
Male
|
20785 (48.2)
|
6760 (63.2)
|
5476 (53.6)
|
4953 (43.9)
|
3596 (33)
|
|
|
Female
|
22309 (51.8)
|
3938 (36.8)
|
4747 (46.4)
|
6323 (56.1)
|
7301 (67)
|
|
|
Age, Mean ± SD
|
49.2 ± 18.0
|
42.3 ± 16.4
|
48.8 ± 17.3
|
52.3 ± 17.8
|
53.1 ± 18.4
|
< 0.001
|
|
Race, n (%)
|
< 0.001
|
|||||
|
Mexican American
|
7358 (17.1)
|
2089 (19.5)
|
1873 (18.3)
|
1827 (16.2)
|
1569 (14.4)
|
|
|
Other Hispanic
|
3399 ( 7.9)
|
722 (6.7)
|
845 (8.3)
|
974 (8.6)
|
858 (7.9)
|
|
|
Non-Hispanic white
|
19762 (45.9)
|
5802 (54.2)
|
4941 (48.3)
|
5029 (44.6)
|
3990 (36.6)
|
|
|
Non-Hispanic black
|
8643 (20.1)
|
1035 (9.7)
|
1565 (15.3)
|
2451 (21.7)
|
3592 (33)
|
|
|
Other race
|
3932 ( 9.1)
|
1050 (9.8)
|
999 (9.8)
|
995 (8.8)
|
888 (8.1)
|
|
|
Marital, n (%)
|
< 0.001
|
|||||
|
Married or living with partners
|
26356 (61.2)
|
6883 (64.3)
|
6598 (64.5)
|
6818 (60.5)
|
6057 (55.6)
|
|
|
Widowed
|
3479 ( 8.1)
|
397 (3.7)
|
642 (6.3)
|
1090 (9.7)
|
1350 (12.4)
|
|
|
Divorced or Separated
|
5923 (13.7)
|
1034 (9.7)
|
1348 (13.2)
|
1713 (15.2)
|
1828 (16.8)
|
|
|
Never married
|
7336 (17.0)
|
2384 (22.3)
|
1635 (16)
|
1655 (14.7)
|
1662 (15.3)
|
|
|
Education, n (%)
|
< 0.001
|
|||||
|
Less than high school
|
11051 (25.6)
|
2383 (22.3)
|
2597 (25.4)
|
2954 (26.2)
|
3117 (28.6)
|
|
|
High school
|
9940 (23.1)
|
2396 (22.4)
|
2311 (22.6)
|
2624 (23.3)
|
2609 (23.9)
|
|
|
College or above
|
22103 (51.3)
|
5919 (55.3)
|
5315 (52)
|
5698 (50.5)
|
5171 (47.5)
|
|
|
PIR, Mean ± SD
|
2.6 ± 1.6
|
2.8 ± 1.6
|
2.7 ± 1.6
|
2.5 ± 1.6
|
2.3 ± 1.6
|
< 0.001
|
|
BMI, Mean ± SD
|
29.0 ± 6.8
|
26.5 ± 4.9
|
28.1 ± 5.7
|
29.5 ± 6.6
|
31.7 ± 8.4
|
< 0.001
|
|
Smoking status, n (%)
|
0.026
|
|||||
|
Never
|
23356 (54.2)
|
5831 (54.5)
|
5553 (54.3)
|
6081 (53.9)
|
5891 (54.1)
|
|
|
Current
|
10759 (25.0)
|
2546 (23.8)
|
2557 (25)
|
2866 (25.4)
|
2790 (25.6)
|
|
|
Former
|
8979 (20.8)
|
2321 (21.7)
|
2113 (20.7)
|
2329 (20.7)
|
2216 (20.3)
|
|
|
Alcohol consumption, n (%)
|
< 0.001
|
|||||
|
Yes
|
5510 (12.8)
|
1033 (9.7)
|
1215 (11.9)
|
1576 (14)
|
1686 (15.5)
|
|
|
No
|
8955 (20.8)
|
1218 (11.4)
|
1743 (17)
|
2584 (22.9)
|
3410 (31.3)
|
|
|
Not recorded
|
28629 (66.4)
|
8447 (79)
|
7265 (71.1)
|
7116 (63.1)
|
5801 (53.2)
|
|
|
DM, n (%)
|
< 0.001
|
|||||
|
No
|
38056 (88.3)
|
10090 (94.3)
|
9305 (91)
|
9833 (87.2)
|
8828 (81)
|
|
|
Yes
|
5038 (11.7)
|
608 (5.7)
|
918 (9)
|
1443 (12.8)
|
2069 (19)
|
|
|
CHD, n (%)
|
< 0.001
|
|||||
|
No
|
41312 (95.9)
|
10468 (97.9)
|
9898 (96.8)
|
10722 (95.1)
|
10224 (93.8)
|
|
|
Yes
|
1782 ( 4.1)
|
230 (2.1)
|
325 (3.2)
|
554 (4.9)
|
673 (6.2)
|
|
|
hypertension, n (%)
|
< 0.001
|
|||||
|
No
|
28336 (65.8)
|
8285 (77.4)
|
7059 (69.1)
|
6998 (62.1)
|
5994 (55)
|
|
|
Yes
|
14758 (34.2)
|
2413 (22.6)
|
3164 (30.9)
|
4278 (37.9)
|
4903 (45)
|
|
|
cancer, n (%)
|
< 0.001
|
|||||
|
No
|
39201 (91.0)
|
10078 (94.2)
|
9402 (92)
|
10113 (89.7)
|
9608 (88.2)
|
|
|
Yes
|
3893 ( 9.0)
|
620 (5.8)
|
821 (8)
|
1163 (10.3)
|
1289 (11.8)
|
3.2. Associations between RAR and cancer
In the multivariable logistic regression analyses, after adjusting for potential confounders (as shown in Table 2, model 3), RAR expressed as a continuous variable (per unit increase) was positively associated with the probability of developing cancer (OR = 1.3, 95% CI = 1.2 ~ 1.39, P < 0.001). Additionally, participants in the fourth quartile (Q4) demonstrated a significantly higher probability of cancer (OR = 2.18, 95% CI = 1.97 ~ 2.41, P < 0.001) when compared to those in the first quartile (Q1), also after adjusting for potential confounders (refer to Table 3, model 3).
|
Variable
|
No.
|
cancer(%)
|
model 1
|
model 2
|
model 3
|
|||
|---|---|---|---|---|---|---|---|---|
|
OR (95%CI)
|
p-value
|
OR (95%CI)
|
p-value
|
OR (95%CI)
|
p-value
|
|||
|
RAR
|
43094
|
3893 (9)
|
1.47 (1.39 ~ 1.56)
|
< 0.001
|
1.34 (1.25 ~ 1.44)
|
< 0.001
|
1.3 (1.2 ~ 1.39)
|
< 0.001
|
|
RAR(quartile)
|
||||||||
|
Q1
|
10698
|
620 (5.8)
|
1(Ref)
|
1(Ref)
|
1(Ref)
|
|||
|
Q2
|
10223
|
821 (8)
|
1.42 (1.27 ~ 1.58)
|
< 0.001
|
1.02 (0.91 ~ 1.14)
|
0.729
|
1.02 (0.91 ~ 1.15)
|
0.725
|
|
Q3
|
11276
|
1163 (10.3)
|
1.87 (1.69 ~ 2.07)
|
< 0.001
|
1.16 (1.04 ~ 1.29)
|
0.008
|
1.15 (1.03 ~ 1.28)
|
0.016
|
|
Q4
|
10897
|
1289 (11.8)
|
2.18 (1.97 ~ 2.41)
|
< 0.001
|
1.41 (1.26 ~ 1.58)
|
< 0.001
|
1.36 (1.22 ~ 1.53)
|
< 0.001
|
|
Trend test
|
< 0.001
|
< 0.001
|
< 0.001
|
|||||
|
Subgroup
|
No.of participants
N
|
Cancer
n(%)
|
OR(95%CI)
|
P for interaction
|
|---|---|---|---|---|
|
Gender
|
< 0.001
|
|||
|
Male
|
20785
|
1849 (8.9)
|
1.34 (1.2 ~ 1.5)
|
|
|
Female
|
22309
|
2044 (9.2)
|
1.13 (1.02 ~ 1.25)
|
|
|
Age(years)
|
0.089
|
|||
|
< 65
|
32923
|
1559 (4.7)
|
1.25 (1.12 ~ 1.39)
|
|
|
≥ 65
|
10171
|
2334 (22.9)
|
1.26 (1.14 ~ 1.39)
|
|
|
BMI(kg/m2)
|
0.219
|
|||
|
< 25
|
12728
|
1123 (8.8)
|
1.2 (1.05 ~ 1.38)
|
|
|
25–30
|
14582
|
1375 (9.4)
|
1.35 (1.18 ~ 1.55)
|
|
|
≥ 30
|
15784
|
1395 (8.8)
|
1.3 (1.16 ~ 1.47)
|
|
|
Race
|
0.070
|
|||
|
Mexican American
|
7358
|
268 (3.6)
|
1 (0.77 ~ 1.31)
|
|
|
Other Hispanic
|
3399
|
186 (5.5)
|
1.33 (1 ~ 1.77)
|
|
|
Non-Hispanic white
|
19762
|
2777 (14.1)
|
1.26 (1.14 ~ 1.39)
|
|
|
Non-Hispanic black
|
8643
|
505 (5.8)
|
1.39 (1.19 ~ 1.61)
|
|
|
Other race
|
3932
|
157 (4)
|
1.65 (1.18 ~ 2.29)
|
|
|
Education
|
0.224
|
|||
|
Less than high school
|
11051
|
857 (7.8)
|
1.28 (1.12 ~ 1.47)
|
|
|
High school
|
9940
|
902 (9.1)
|
1.18 (1 ~ 1.38)
|
|
|
College or above
|
22103
|
2134 (9.7)
|
1.36 (1.23 ~ 1.51)
|
|
|
Smoking status
|
0.663
|
|||
|
Never
|
23356
|
1706 (7.3)
|
1.26 (1.12 ~ 1.42)
|
|
|
Current
|
10759
|
1576 (14.6)
|
1.34 (1.19 ~ 1.52)
|
|
|
Former
|
8979
|
611 (6.8)
|
1.29 (1.11 ~ 1.51)
|
|
|
Alcohol consumption
|
0.602
|
|||
|
Yes
|
5510
|
458 (8.3)
|
1.19 (0.96 ~ 1.49)
|
|
|
No
|
8955
|
943 (10.5)
|
1.27 (1.1 ~ 1.45)
|
|
|
Not recorded
|
28629
|
2492 (8.7)
|
1.32 (1.2 ~ 1.45)
|
The fully adjusted analyses using restricted cubic splines (RCS) suggested a J-shaped nonlinear relationship (P for nonlinearity = 0.026) between RAR and cancer. A significant positive association was observed, with the slope of the curve becoming steeper at an RAR of 2.941 mmol/L. In contrast, for RAR values below 2.941 mmol/L, the estimated dose-response curve closely approximated a horizontal line, indicating that the relationship between RAR and cancer risk was not statistically significant (P > 0.05). Furthermore, cancer incidence increased with rising RAR levels up to the identified turning point, with the OR for cancer being 1.248 (95% CI: 1.133 ~ 1.374) as presented in Fig. 2, Table 4.
Fig. 2
Adjusted Relationship between RAR and Cancer Odds Ratio. Solid and dashed lines represent the predicted value and 95% confidence intervals. They were adjusted for age, gender, race, marital, gender education level, family income level, body mass index (BMI), smoking status, alcohol drinking status, hypertension, diabetes and coronary heart disease.
|
RAR
|
Adjusted OR
|
95% CI
|
P value
|
|---|---|---|---|
|
< 2.941
|
0.869
|
0.511 ~ 1.477
|
0.6033
|
|
≥ 2.941
|
1.248
|
1.133 ~ 1.374
|
< 0.001
|
3.3. Stratified analyses based on additional variables
Stratified analysis was performed across various subgroups to assess potential differences in how RAR relates to cancer. We conducted stratified analyses based on the variables gender, age, BMI, race, education, smoking status, and drinking status, as detailed in Fig. 3, Table 3. In the subgroup stratified by gender, we identified significant interaction effects (P < 0.001).
Fig. 3
The relationship between RAR and cancer according to basic features. Except for the stratification component itself, each stratification factor was adjusted for all other variables (age, gender, race, marital, gender education level, family income level, body mass index (BMI), smoking status, alcohol drinking status, hypertension, diabetes and coronary heart disease).
4. Discussion
This large-scale cross-sectional survey found a nonlinear relationship (J-shaped) between RAR and the risk of cancer in US adults.When RAR was more than 2.941 mmol/L, the risk of cancer rises significantly, with a corresponding OR value of 1.248 (95% CI: 1.133–1.374). However, no significant association was seen between RAR and cancer risk when RAR was less than 2.941 mmol/L. The association between RAR and cancer persisted, according to stratified analyses. Subgroup analysis indicates a gender interaction (P < 0.001), suggesting that RAR may differ in its impact on cancer risk between males and females, highlighting the need for further research to explore the underlying causes.
Existing literature supports inflammation-nutrition biomarkers in cancer prognosis, yet RAR's predictive value for prevalence remains underexplored. For instance, the neutrophil-to-lymphocyte ratio (NLR), a marker of systemic inflammation, has been linked to higher cancer risk and poorer prognoses. In colorectal cancer patients, a one-unit preoperative NLR increase is tied to a 12% drop in overall survival (HR = 1.12, 95% CI: 1.03–1.21)[26]. Li et al. found that NLR is independently associated with a 20% higher cancer risk (OR = 1.2, 95% CI: 1.05–1.36)[27]. Another study[28] shows that cancer patients have significantly higher NLR and PLR levels than controls (5.30 vs. 2.60, P < 0.001; 217 vs. 136, P < 0.001), indicating that NLR and PLR may serve as novel diagnostic biomarkers for non-small cell lung cancer. The albumin-to-globulin ratio (AGR), reflecting nutritional status, also correlates with cancer outcomes. Low preoperative AGR often signals malnutrition and impaired immunity, and is associated with poorer postoperative survival in colorectal cancer patients, with a risk ratio as high as 3.04 (95% CI: 2.06–4.49)[29]. While C-reactive protein-to-albumin ratio (CAR) is linked to both inflammation and nutrition, most CAR research focuses on prognosis assessment[30–32].
RAR, integrating both inflammatory and nutritional factors, may outperform single indicators in predicting cancer risk. Its "J"-shaped curve suggests a complex physiological mechanism potentially tied to red blood cell oxygen-carrying function impacting the tumor microenvironment, albumin maintaining colloid osmotic pressure, and immune regulation. This underscores RAR’s unique value as a novel biomarker. RAR research is still in its infancy, but some initial findings are promising. One study[33] found a nonlinear relationship between RAR and colorectal cancer, aligning with our results. Multivariate regression showed that higher RAR levels are significantly linked to increased colorectal cancer risk (OR = 1.48, 95% CI = 1.04–2.11), with a nonlinear relationship between RAR and colorectal cancer risk (P = 0.0458). In our study, multivariate regression indicated that each RAR unit increase raises cancer risk by 47% (OR = 1.47, 95% CI = 1.39–1.56). Individuals in the highest RAR tertile had an 118% higher cancer prevalence than those in the lowest tertile (OR = 2.18, 95% CI = 1.97–2.41), with consistent results after full model adjustment. We also observed a J-shaped nonlinear relationship between RAR and cancer risk when it was more than 2.941 mmol/L, while there is no statistically significant association between RAR and cancer risk before the 2.941 mmol/L threshold. Additionally, RAR has been explored for predicting cancer patient survival. One study[34] found elevated RAR levels to be associated with a higher risk of malignant neoplasm-specific mortality. Another study[35] demonstrated that high RAR is independently linked to higher all-cause mortality in cancer patients, with a 74% increased risk in the high RAR group (HR = 1.74, 95% CI = 1.48–2.04).
Despite these clinically significant findings, several limitations exist. First, residual confounding from unmeasured or unknown factors cannot be fully ruled out, despite regression models and stratified analyses. Second, the findings are based solely on U.S. adults and may not be generalizable to other populations. Third, cancer type heterogeneity might confound the effect estimates. Different cancers have distinct mechanisms, prognostic factors, and biomarker relationships, yet this study lacks detailed cancer subtype stratification. Future research should focus on specific cancers to clarify RAR’s cancer type-specific effects. Finally, the cross-sectional design precludes establishing causality between RAR and cancer risk, necessitating further longitudinal research for confirmation.
Our study found that increased RAR was significantly associated with higher cancer risk in adults, highlighting its potential as a practical biomarker for risk assessment. As RAR can be measured routinely, it offers a simple, reliable, and cost-effective tool for identifying high-risk individuals in clinical practice. Future prospective studies are needed to confirm these findings and to clarify the mechanisms linking RAR to cancer development and progression.
Acknowledgments:
We thank the National health and Nutrition examination survey participants, staff, and National center for health statistics for their valuable contributions.
Declaration of Interest
statement:
The authors report no conflict of interest.
Author Contributions statement:
Qian Wang: Conceptualization, Methodology, Software, Formal analysis, Data curation, Validation, Investigation, Visualization, Writing - original draft, Writing - review & editing. The author confirms sole responsibility for the entire content of this manuscript.
A
Data Availability
The datasets generated during the current study are available in the NHANES repository (https:// www.cdc.gov/nchs/nhanes/).
A
Funding:
This study did not receive any funding in any form.
A
Author Contribution
Qian Wang: Conceptualization, Methodology, Software, Formal analysis, Data curation, Validation, Investigation, Visualization, Writing - original draft, Writing - review & editing. The author confirms sole responsibility for the entire content of this manuscript.
Reference
1.
Bray, F. et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. Cancer J. Clin. 74 (3), 229–263. https://doi.org/10.3322/caac.21834 (2024).
2.
Ko, C. A. et al. Prognostic Value of Neutrophil Percentage-to-Albumin Ratio in Patients with Oral Cavity Cancer. Cancers 14 (19), 4892. https://doi.org/10.3390/cancers14194892 (2022).
3.
Wang, M. et al. Associations between neutrophil percentage-to-albumin ratio with all-cause and cause-specific mortality among US cancer survivors: evidence from NHANES 2005–2018. Front. Nutr. 12, 1541609. https://doi.org/10.3389/fnut.2025.1541609 (2025).
4.
Zhao, B., Guo, H., Wu, W. & Duan, G. Hemoglobin, albumin, lymphocyte and platelet (HALP) score can predict the prognosis of patients with non-small cell lung cancer (NSCLC). Asian J. Surg. 46 (11), 4891–4892. https://doi.org/10.1016/j.asjsur.2023.05.152 (2023).
5.
Xu, H., Zheng, X., Ai, J. & Yang, L. Hemoglobin, albumin, lymphocyte, and platelet (HALP) score and cancer prognosis: A systematic review and meta-analysis of 13,110 patients. Int. Immunopharmacol. 114, 109496. https://doi.org/10.1016/j.intimp.2022.109496 (2023).
6.
Jiang, T. et al. Prognostic significance of hemoglobin, albumin, lymphocyte, and platelet (HALP) score in breast cancer: a propensity score-matching study. Cancer Cell Int. 24 (1), 230. https://doi.org/10.1186/s12935-024-03419-w (2024).
7.
Yang, M. et al. Association between C-reactive protein-albumin-lymphocyte (CALLY) index and overall survival in patients with colorectal cancer: From the investigation on nutrition status and clinical outcome of common cancers study. Front. Immunol. 14, 1131496. https://doi.org/10.3389/fimmu.2023.1131496 (2023).
8.
Jia, P. et al. Association between C-reactive protein-albumin-lymphocyte index and overall survival in patients with esophageal cancer. Clin. Nutr. 45, 212–222. https://doi.org/10.1016/j.clnu.2024.12.032 (2025).
9.
Wang, X. & Wang, Y. The prognostic nutritional index is prognostic factor of gynecological cancer: A systematic review and meta-analysis. Int. J. Surg. (London England). 67, 79–86. https://doi.org/10.1016/j.ijsu.2019.05.018 (2019).
10.
Yamamoto, T., Kawada, K. & Obama, K. Inflammation-Related Biomarkers for the Prediction of Prognosis in Colorectal Cancer Patients. Int. J. Mol. Sci. 22 (15), 8002. https://doi.org/10.3390/ijms22158002 (2021).
11.
Zheng, Y. et al. Prognostic value of a baseline prognostic nutritional index for patients with prostate cancer: a systematic review and meta-analysis. Prostate Cancer Prostatic Dis. 27 (4), 604–613. https://doi.org/10.1038/s41391-023-00689-9 (2024).
12.
Zhao, L. et al. Association of prognostic nutritional index with mortalities in American adult cancer survivors: A cohort study based on NHANES, 1999–2018. Food Sci. Nutr. 12 (3), 1834–1846. https://doi.org/10.1002/fsn3.3877 (2023).
13.
Luan, C. W. et al. Prognostic Value of C-Reactive Protein-to-Albumin Ratio in Head and Neck Cancer: A Meta-Analysis. Diagnostics (Basel Switzerland). 11 (3), 403. https://doi.org/10.3390/diagnostics11030403 (2021).
14.
Liao, C. K. et al. Prognostic value of the C-reactive protein to albumin ratio in colorectal cancer: an updated systematic review and meta-analysis. World J. Surg. Oncol. 19 (1), 139. https://doi.org/10.1186/s12957-021-02253-y (2021).
15.
Fagarasan, G. et al. The Value of Preoperative C-Reactive Protein to Albumin Ratio as a Prognostic Biomarker in Colon Cancer Patients. Med. (Kaunas Lithuania). 60 (7), 1054. https://doi.org/10.3390/medicina60071054 (2024).
16.
Coussens, L. M. & Werb, Z. Inflammation and cancer. Nature 420 (6917), 860–867. https://doi.org/10.1038/nature01322 (2002).
17.
Mantovani, A., Allavena, P., Sica, A. & Balkwill, F. Cancer-related inflammation. Nature 454 (7203), 436–444. https://doi.org/10.1038/nature07205 (2008).
18.
Arends, J. et al. ESPEN guidelines on nutrition in cancer patients. Clinical nutrition (Edinburgh. Scotland) 36 (1), 11–48. https://doi.org/10.1016/j.clnu.2016.07.015 (2017).
19.
García-Escobar, A. et al. Red Blood Cell Distribution Width is a Biomarker of Red Cell Dysfunction Associated with High Systemic Inflammation and a Prognostic Marker in Heart Failure and Cardiovascular Disease: A Potential Predictor of Atrial Fibrillation Recurrence. High. blood Press. Cardiovasc. prevention: official J. Italian Soc. Hypertens. 31 (5), 437–449. https://doi.org/10.1007/s40292-024-00662-0 (2024).
20.
Hong, J. et al. Impact of red cell distribution width and red cell distribution width/albumin ratio on all-cause mortality in patients with type 2 diabetes and foot ulcers: a retrospective cohort study. Cardiovasc. Diabetol. 21 (1), 91. https://doi.org/10.1186/s12933-022-01534-4 (2022).
21.
Xiao, J. et al. Association between red cell distribution width and all-cause mortality in patients with breast cancer: A retrospective analysis using MIMIC-IV 2.0. PloS one. 19 (5), e0302414. https://doi.org/10.1371/journal.pone.0302414 (2024).
22.
Gupta, D. & Lis, C. G. Pretreatment serum albumin as a predictor of cancer survival: a systematic review of the epidemiological literature. Nutr. J. 9, 69. https://doi.org/10.1186/1475-2891-9-69 (2010).
23.
Ding, J., Zhang, Y. & Chen, X. Red cell distribution width to albumin ratio is associated with asthma risk: a population-based study. Front. Med. https://doi.org/10.3389/fmed.2024.1493463 (2024). 11,1493463.
24.
Chen, J., Zhang, D., Zhou, D., Dai, Z. & Wang, J. Association between red cell distribution width/serum albumin ratio and diabetic kidney disease. J. diabetes. 16 (7), e13575. https://doi.org/10.1111/1753-0407.13575 (2024).
25.
Xu, Y. & Li, S. J-Shaped relationship between the red cell distribution width to albumin ratio and erectile dysfunction: a cross-sectional study from NHANES 2001–2004. Front. Endocrinol. 16, 1545272. https://doi.org/10.3389/fendo.2025.1545272 (2025).
26.
Naszai, M., Kurjan, A. & Maughan, T. S. The prognostic utility of pre-treatment neutrophil-to-lymphocyte-ratio (NLR) in colorectal cancer: A systematic review and meta-analysis. Cancer Med. 10 (17), 5983–5997. https://doi.org/10.1002/cam4.4143 (2021).
27.
Li, G. P. et al. Association between the neutrophil-to-lymphocyte ratio and cancer in adults from NHANES 2005–2018: a cross-sectional study. Sci. Rep. 14 (1), 23678. https://doi.org/10.1038/s41598-024-75252-0 (2024).
28.
Budin, C. E. et al. Neutrophil-to-Lymphocyte and Platelet-to-Lymphocyte Ratio: Side by Side with Molecular Mutations in Patients with Non-Small Cell Lung Cancer-The INOLUNG Study. Cancers 16 (16), 2903. https://doi.org/10.3390/cancers16162903 (2024).
29.
Li, K. et al. Preoperative pan-immuno-inflammatory values and albumin-to-globulin ratio predict the prognosis of stage I-III colorectal cancer. Sci. Rep. 15 (1), 11517. https://doi.org/10.1038/s41598-025-96592-5 (2025).
30.
Takemoto, Y. et al. Preoperative High C-Reactive Protein to Albumin Ratio Predicts Short- and Long-Term Postoperative Outcomes in Elderly Gastric Cancer Patients. Cancers 16 (3), 616. https://doi.org/10.3390/cancers16030616 (2024).
31.
Luan, C. W. et al. Prognostic Value of C-Reactive Protein-to-Albumin Ratio in Head and Neck Cancer: A Meta-Analysis. Diagnostics (Basel Switzerland). 11 (3), 403. https://doi.org/10.3390/diagnostics11030403 (2021).
32.
Hajibandeh, S. et al. Preoperative C-Reactive Protein-to-Albumin Ratio and Its Ability to Predict Outcomes of Pancreatic Cancer Resection: A Systematic Review. Biomedicines, 11(7), 1983. (2023). https://doi.org/10.3390/biomedicines11071983
33.
Luo, J., Zhu, P. & Zhou, S. Association between the red blood cell distribution width-to-albumin ratio and risk of colorectal and gastric cancers: a cross-sectional study using NHANES 2005–2018. BMC Gastroenterol. 25 (1), 316. https://doi.org/10.1186/s12876-025-03871-6 (2025).
34.
Hao, M. et al. Ratio of Red Blood Cell Distribution Width to Albumin Level and Risk of Mortality. JAMA Netw. open. 7 (5), e2413213. https://doi.org/10.1001/jamanetworkopen.2024.13213 (2024).
35.
Lu, C. et al. Red blood cell distribution width-to-albumin ratio is associated with all-cause mortality in cancer patients. J. Clin. Lab. Anal. 36 (5), e24423. https://doi.org/10.1002/jcla.24423 (2022).
Note
Multivariate logistic regression models with three models to control for confounding factors.
Model 1: Unadjusted.
Model 2:Adjusted for sociodemographics (age, gender, race, marital, gender education level and PIR).
Model 3 Adjusted for Model 2 + clinical factors (BMI, smoking, alcohol, hypertension, diabetes and CHD).
Abbreviations:OR: odds ratio; CI: confidence interval.
Notes
adjusted for age, gender, race, marital, gender education level, family income level, body mass index (BMI),smoking status, alcohol drinking status, hypertension, diabetes and coronary heart disease.
Note
adjusted for age, gender, race, marital, gender education level, family income level, body mass index (BMI), smoking status, alcohol drinking status, hypertension, diabetes and coronary heart disease.
Abbreviations:OR: odds ratio; CI: confidence interval.