ABSTRACT
Background
Inflammation‐related biomarkers have been implicated in a wide range of cancer‐related symptoms and patient‐reported outcomes (PROs), including fatigue, pain, and depression. To interpret biomarker data among cancer patients, normative reference values from healthy populations are essential.
Objective
This cross‐sectional study aimed to provide sex‐ and age‐stratified reference data distributions for 12 inflammation‐related serum biomarkers in a representative sample of Dutch adults without a history of cancer.
Methods
Adults from the LISS panel were selected to match cancer cohorts from the PROFILES registry in age and sex. Participants completed questionnaires and optionally provided blood samples using harmonized protocols aligned with the PROFILES registry. Serum levels of IL‐1α, IL‐1β, IL‐6, IL‐8, IL‐10, IL‐17A, IL‐22, CRP, TGF‐α, IFN‐γ, IL‐1RA, and soluble TNF receptors I and II (sTNFRI/II) were quantified using the Meso Scale Discovery (MSD) platform. Cytokine levels were stratified by demographics, lifestyle factors, and self‐reported health conditions. ANOVAs were used to compare groups.
Results
In total 720 panel members filled out the questionnaires, of which 265 (mean age = 59.0, 47.9% male) provided additional serum blood samples. Biomarker levels varied by sex, age, BMI, smoking status, and number of health conditions. Notably, higher levels of IL‐1RA and sTNFRI/II were observed in individuals with obesity or diabetes. Individuals with osteoarthritis exhibited elevated IL‐6 and IL‐1β levels.
Conclusion
This study offers a robust reference dataset for key inflammatory biomarkers, stratified by relevant demographics, lifestyle factors, and self‐reported health conditions. These data facilitate interpretation of biomarker–PRO relationships in cancer survivors and support comparative research across chronic disease populations.
Keywords: biomarkers, general population, inflammation, reference distribution
1. Introduction
Improvements in early detection and treatment—alongside a growing aging population—have led to a significant rise in the number of cancer survivors. Today, more than 50 million people worldwide are living with a cancer diagnosis received within the past 5 years [1]. While survival rates have improved, many of these individuals continue to experience a wide range of physical and psychosocial difficulties stemming from cancer and its treatment [2]. In response, the field of cancer survivorship has expanded considerably over recent decades, aiming to better understand and address these issues.
Fatigue, pain, depression, and reduced quality of life (QoL) are common issues among cancer patients and survivors, significantly impacting daily functioning [3]. There is increasing evidence supporting a biological foundation for these subjective experiences [4, 5, 6, 7]. Several studies suggest that immune dysregulation plays a key role in the development and persistence of patient‐reported outcomes (PROs) in oncological populations [7, 8, 9, 10]. Particularly elevated concentrations of pro‐inflammatory cytokines such as interleukin‐6 (IL‐6), IL‐1β, and tumor necrosis factor‐alpha (TNF‐α) are involved in poor PROs such as fatigue, pain, and depressive symptoms among cancer patients [7, 8, 9, 10].
Unraveling the biological mechanisms underlying impaired PROs is crucial for developing new treatments minimizing adverse effects of cancer and its treatment thereby preserving QoL. To accurately assess the impact of cancer (treatment) on PROs among cancer survivors, it is essential to compare these biological associations with those observed in a healthy reference population. Only then can the specific effects of cancer and its treatment be distinguished from normal aging or health conditions [3, 6].
An important strategy in this context is the establishment of reference distributions for relevant inflammatory markers in individuals without cancer, stratified by known associates of pro‐inflammatory markers—that is, age [11], sex [12], body mass index (BMI) [13], smoking status [14], and health conditions [15]. Such data can provide a necessary reference frame to quantify deviations in cancer survivors and may help explain differences in PROs across subgroups. In this paper, we therefore report concentrations of pro‐inflammatory cytokines in a representative sample of Dutch adults without cancer, stratified by demographic (sex and age) and clinical characteristics (BMI, smoking, and health conditions). These data aim to support future investigations into the biological underpinnings of fatigue, pain, depression, QoL, and other PROs among cancer patients and survivors.
2. Methods
2.1. Setting and Study Population
For this cross‐sectional study normative data were obtained from the Health and Health Complaints study, administered in the Longitudinal Internet studies for the Social Sciences (LISS panel) (www.lissdata.nl). The LISS panel, administered by Centerdata (Tilburg University, the Netherlands), is an online household panel, which comprises approximately 5000 households representative of the Dutch‐speaking adult population in the Netherlands. The panel is based on a traditional probability sample drawn from the Dutch population register by Statistics Netherlands [16].
To study long‐term and late effects of cancer and its treatment, including changes in patient‐reported outcomes (PROs) and recovery trajectories, the PROFILES (Patient Reported Outcomes Following Initial treatment and Long‐term Evaluation of Survivorship) registry was established in 2009 [17]. Facilitating research on the impact of cancer and its treatment, members of the LISS panel (formally known as Centerpanel, 2009–2014) reported annually on similar PROs like the patients in PROFILES, enabling a comparison beyond that of normal aging and health conditions [6].
PROFILES obtained funding to expand this normative cohort with the collection of blood samples; details can be found elsewhere [18]. In the autumn of 2023, LISS panel members were selected based on age and sex to match our PROFILES cancer cohorts. Additional selection criteria included residence within a 17‐min driving radius of one of the 16 hospitals involved in the ongoing PROFILES studies—WaTCh [19] and OPTIMUM [20]. This matched population was included following medical ethical approval granted by the Medical Ethics Review Committee (METC) (WaTCh: NL65161.028.18 / P1838; OPTIMUM: NL66913.028.18) and approval by all participating hospitals [18].
2.2. Blood Collection and Processing
The stratified subsample received an information letter and a blood collection package, with a request to visit a local hospital for venipuncture. The package included labeled blood tubes (BD Vacutainer 1 clot tube and 1 EDTA tube) and cryovials, a laboratory requisition form, a brief questionnaire to control for acute pre‐analytical influences that may affect biomarker concentrations (e.g., recent caffeine consumption, physical activity, acute illness, and vaccination preceding blood collection) and instructions for the lab technician. These blood packages were made by researchers and students in a cleaned room (i.e., hypoallergenic) to prevent contamination.
All blood samples were processed uniformly across all sites, following the same standard operating procedure used for cancer patients [18]. Blood was collected using both one clot and one EDTA tube. Serum obtained from the clot tube was used for all biomarker analyses reported in this study. Clot tubes were allowed to clot at room temperature (for a maximum of 30 min) and subsequently centrifuged at 1800 × g for 10 min at 20°C with brake. The resulting serum was carefully separated, aliquoted into labeled cryovials, and temporarily stored at room temperature prior to long‐term storage at −80°C. All participating laboratories were equipped with standardized facilities. Each sample was labeled with a QR code containing metadata, including study identifier, collection site, participant number, and sample type (e.g., serum, plasma, buffy coat, or whole blood). After sample collection was completed, all materials were transported on dry ice at approximately −80°C to a central biobank within a 24‐h time frame.
Participation in blood sample collection was optional and served as a supplement to the primary collection of PRO data. Participants were offered €35 compensation for their time and effort to help reduce participation burden and improve inclusion rates. Additionally, they were given the opportunity to consult with an independent physician regarding study participation and could contact the LISS panel helpdesk during office hours for further assistance. All participants provided informed consent for processing and storing of their data.
2.3. Measurements
Biomarkers were measured in the collected blood specimens and selected based on their established relevance and utility in relation to PROs. For each participant, one serum aliquot was thawed for analysis, while the remaining aliquots remained stored to facilitate future investigations.
A panel of 12 inflammation‐related biomarkers was selected for quantification in undiluted serum samples: Interleukin‐1 alpha (IL‐1α), IL‐1 beta (IL‐1β), IL‐6, IL‐8, IL‐10, IL‐17A, IL‐22, C‐Reactive Protein (CRP), Transforming Growth Factor‐alpha (TGF‐α), Interferon gamma (IFN‐γ), and soluble Tumor Necrosis Factor Receptors 1 and 2 (sTNFRI and sTNFRII). Biomarker concentrations were measured in serum using commercially available single‐ or multiplex sandwich immunoassay kits on the Meso Scale Discovery (MSD) platform (Rockville, MD, USA), including the appropriate MSD V‐PLEX panels with manufacturer‐specified lower limits of detection; details are provided in the appendix. Serum samples were defrosted at room temperature with shaking and centrifuged again at 2000 RPM to obtain the soluble fraction prior to measurement. Measurements were performed according to the manufacturer's protocol and analyzed using the MESO QuickPlex SQ 120MM system [21].
Sociodemographic and lifestyle factors collected included age, sex, living situation (with a partner and/or children), educational level, income, and employment status. Current smoking status, (former, never, currently) and alcohol use were self‐reported via questionnaires. Participants were classified into occasional (0–14 g/day), moderate (14–41.9 g/day), or heavy (≥ 42 g/day) drinkers following the National institute of Alcohol Abuse and Alcoholism (NIAA) [22]. BMI was calculated as weight in kilograms divided by their squared height in meters and categorized into healthy weight (18–24.9 kg/m2), overweight (25–29.9 kg/m2), or obese (≥ 30 kg/m2).
Self‐reported health conditions were assessed using an adapted version of the Self‐administered Comorbidity Questionnaire (SCQ) [23]. Respondents were asked whether they currently had, or had experienced in the past 12 months, any of the following conditions: asthma/COPD, depression, diabetes mellitus, heart disease, hypertension, osteoarthritis, rheumatoid arthritis, stroke, stomach, kidney, liver, or thyroid disease. Information on prior cancer diagnoses was obtained through a separate question.
2.4. Statistics
Participants who reported to have ever been diagnosed with cancer were excluded from the sample. Routinely collected sociodemographic data from the LISS panel allowed for a comparison between respondents and non‐respondents, employing Student's t‐tests for continuous variables and chi‐squared tests for categorical variables.
As is common in the measurement of the included biomarkers, there were non‐detectable values, which were imputed using multiple imputations with a minimum of zero and a maximum of the lowest detected value for each biomarker. One thousand iterations ensured model convergence, minimizing bias and stabilizing variance. Multiple imputation was preferred over substitution methods (e.g., LOD/2), as it better preserves variance and reduces bias in skewed biomarker distributions. Outliers were all within the set cut‐off (1.5 times interquartile range). As a sensitivity analysis, descriptive statistics were compared between the imputed and non‐imputed datasets. Biomarker means and standard deviations were highly similar across both approaches, indicating that multiple imputations had negligible influence on low‐level cytokine values and did not materially affect the lower tail of the distribution.
Mean scores per biomarker were stratified by age group (all ages, 18–29, 30–39, 40–49, 50–59, 60–69, and 70–92) for men and women separately. Other subgroup analyses (smoking status, BMI, and number of health conditions) were conducted across the total sample and should be interpreted separately. The number of health conditions was calculated and categorized into no health condition, one health condition, or two or more health conditions. Biomarker scores were reported for each health condition separately and by the categorized number of health conditions.
Differences across all groups (sex‐stratified age groups; smoking status; BMI; [number of] health conditions) were explored by means of ANOVA's to assist interpretation of observed distributional patterns for exploratory and descriptive purposes only. The primary focus is on the presentation of reference distributions, rather than to establish clinical reference intervals or statistically testing group differences. All tests were performed with SPSS version 29, with a significance level of 5% given the exploratory nature of this study.
3. Results
3.1. Respondents
In total 1027 panel members were invited; of those, 813 responded. Although not intended, 93 respondents were previously diagnosed with cancer (not including basal cell carcinoma, BCC) and thus should not have been included in the study, leaving a total of 720 participating panel members. Of those participants, 36.8% (n = 265) had their blood drawn in addition to their questionnaire response.
Compared to participating panel members without a blood sample (n = 455), those who completed the questionnaires and provided a blood sample (n = 265) smoked less frequently but reported more health conditions, such as high blood pressure.
3.2. Sample Characteristics
The 265 respondents had a mean age of 59 years, and 47.9% were male. Across the total sample—and similarly for both men and women—about 25% were under 50 years old, ~20% were 50–59, and ~26%–32% were aged 60–69 or 70–88 (Table 1). Most participants reported a healthy or slightly elevated weight and had never smoked. Nearly 75% reported at least one health condition, with hypertension being the most common. Depression and osteoarthritis were reported more frequently by women (all p's < 0.05).
TABLE 1.
Participant characteristics in numbers and percentages, with age in mean years (SD).
| Total (n = 265) | Men (n = 127, 47.9%) | Women (n = 138, 52.1%) | X 2 or t‐test | |
|---|---|---|---|---|
| Mean age (years) | 59.01, SD = 14.97 | 60.04, SD = 14.71 | 58.07, SD = 15.20 | t = 1.07, p = 0.28 |
| Age category | 2.21, p = 0.82 | |||
| 18–29 | 13 (4.9) | 5 (3.9) | 8 (5.8) | |
| 30–39 | 24 (9.1) | 12 (9.4) | 12 (8.7) | |
| 40–49 | 26 (9.8) | 11 (8.7) | 15 (10.9) | |
| 50–59 | 56 (21.1) | 24 (18.9) | 32 (23.2) | |
| 60–69 | 70 (26.4) | 35 (26.7) | 35 (26.1) | |
| 70–88 | 76 (28.7) | 40 (31.5) | 36 (26.1) | |
| Living situation | 0.173, p = 0.98 | |||
| Married/co‐habiting | 144 (54.3) | 70 (55.1) | 74 (53.6) | |
| Divorced/separated | 42 (15.8) | 19 (15.0) | 23 (16.7) | |
| Widowed | 13 (4.9) | 6 (4.7) | 7 (5.1) | |
| Never married | 66 (24.9) | 32 (25.2) | 34 (24.6) | |
| Education | 2.20, p = 0.33 | |||
| Low | 56 (21.1) | 24 (18.9) | 32 (23.2) | |
| Middle | 92 (34.7) | 41 (32.2) | 51 (37.0) | |
| High | 117 (44.2) | 62 (48.8) | 55 (39.9) | |
| Employment | ||||
| Paid work (yes) | 128 (48.3) | 69 (54.3) | 59 (42.8) | 3.55, p = 0.06 |
| BMI | 1.29, p = 0.53 | |||
| Healthy weight | 111 (41.9) | 49 (38.6) | 62 (44.9) | |
| Overweight | 111 (41.9) | 55 (43.3) | 56 (40.6) | |
| Obese | 43 (16.2) | 23 (18.1) | 20 (14.5) | |
| Smoking status | 24 (9.1) | 10 (7.9) | 14 (10.1) | 0.51, p = 0.78 |
| Never | 126 (47.5) | 60 (47.2) | 66 (47.8) | |
| Former | 115 (43.4) | 57 (44.9) | 58 (42.0) | |
| Current | 24 (9.1) | 10 (7.9) | 14 (10.1) | |
| Alcohol use (missing 76, 28.7%) | 9.73, p < 0.01 | |||
| Occasional | 136 (72.0) | 61 (62.2) | 75 (82.4) | |
| Moderate | 42 (22.2) | 30 (30.6) | 12 (13.2) | |
| Heavy | 11 (5.8) | 7 (7.1) | 4 (4.4) | |
| Self‐reported health conditions in past 12 months | ||||
|
Number of health‐ conditions |
2.80, p = 0.25 | |||
| 0 | 72 (27.2) | 36 (28.3) | 36 (26.1) | |
| 1 | 59 (22.3) | 33 (26.0) | 26 (18.8) | |
| ≥ 2 | 134 (50.6) | 58 (45.7) | 76 (55.1) | |
| Heart disease | 28 (10.6) | 15 (11.8) | 13 (9.4) | 0.40, p = 0.53 |
| Hypertension | 73 (27.5) | 37 (29.1) | 36 (26.1) | 0.31, p = 0.58 |
| Asthma/COPD | 23 (8.7) | 11 (8.7) | 12 (8.7) | 0.00, p = 0.99 |
| Diabetes | 19 (7.2) | 9 (7.1) | 10 (7.2) | < 0.01, p = 0.96 |
| Depression | 21 (7.9) | 5 (3.9) | 16 (11.6) | 5.31, p = 0.02 |
| Osteoarthritis | 65 (24.5) | 22 (17.3) | 4 (31.2) | 6.84, p < 0.01 |
| Rheumatoid arthritis | 16 (6) | 4 (3.1) | 12 (8.7) | 3.59, p = 0.06 |
Note: BMI was calculated as weight in kilograms divided by their squared height in meters and categorized into healthy weight (18–24.9 kg/m2), overweight (25–29.9 kg/m2), or obese (≥ 30 kg/m2). Alcohol use was categorized into occasional (0–14 g/day), moderate (14–41.9 g/day), or heavy (≥ 42 g/day) drinkers [22].
Abbreviations: %, percentage; N, number; SD = standard deviation. X2, chi‐square test.
3.3. Cytokines by Sex and Age Group
Cytokine levels were calculated separately for men and women and then stratified by age group (Table 2). Results show that sTNFRI levels are higher for older men compared to younger men (F = 2.51, p = 0.03), with a similar finding for sTNFRII for older women compared to younger women (F = 2.93, p = 0.02).
TABLE 2.
Biomarker levels stratified by sex and age groups in means and standard deviations.
| Biomarker | Men | Women | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total n = 127 | 18–29 (n = 5) | 30–39 (n = 12) | 40–49 (n = 11) | 50–59 (n = 24) | 60–69 (n = 35) | 70–88 (n = 40) | Total n = 138 | 18–29 (n = 8) | 30–39 (n = 12) | 40–49 (n = 15) | 50–59 (n = 32) | 60–69 (n = 35) | 70–88 (n = 36) | |
| CRP (mg/ml) | 2.88 (4.35) | 0.17 (0.09) | 1.91 (2.14) | 3.71 (5.35) | 2.35 (3.03) | 2.45 (2.94) | 3.98 (6.04) | 4.41 (7.91) | 2.65 (3.28) | 4.41 (8.53) | 7.16 (16.89) | 3.97 (5.35) | 4.30 (6.15) | 4.17 (6.24) |
| TGF‐α (pg/ml) | 10.73 (5.29) | 10.11 (3.03) | 9.64 (3.77) | 10.55 (4.74) | 10.64 (4.76) | 10.49 (5.80) | 11.44 (5.98) | 12.74 (6.83) | 12.77 (5.57) | 14.81 (6.99) | 13.81 (9.61) | 12.50 (6.21) | 13.25 (7.61) | 11.25 (5.33) |
| IL‐22 (pg/ml) | 0.95 (1.52) | 0.22 (0.21) | 0.75 (1.56) | 1.77 (2.07) | 0.69 (1.11) | 0.74 (1.30) | 1.24 (1.75) | 1.80 (8.92) | 0.69 (1.70) | 2.79 (7.76) | 0.42 (0.53) | 2.83 (14.70) | 0.46 (0.39) | 2.65 (9.67) |
| IL‐17A (pg/ml) | 0.91 (1.30) | 0.63 (0.14) | 0.88 (0.61) | 0.48 (0.28) | 1.26 (2.61) | 0.76 (0.71) | 0.98 (0.82) | 0.99 (0.95) | 0.79 (1.02) | 1.22 (0.93) | 1.06 (1.44) | 0.84 (0.70) | 0.71 (0.43) | 1.32 (1.16) |
| IL‐1 α (pg/ml) | 0.11 (0.47) | 0.13 (0.21) | 0.04 (0.04) | 0.05 (0.04) | 0.20 (0.75) | 0.06 (0.11) | 0.14 (0.58) | 0.11 (0.61) | 0.15 (0.32) | 0.10 (0.09) | 0.03 (0.01) | 0.27 (1.25) | 0.08 (0.18) | 0.05 (0.06) |
| IFN‐γ (pg/ml) | 7.95 (13.47) | 4.40 (2.20) | 5.25 (3.46) | 19.70 (36.95) | 6.65 (8.70) | 5.91 (4.76) | 8.54 (10.96) | 9.62 (18.76) | 7.91 (6.92) | 22.64 (57.08) | 8.05 (12.49) | 5.13 (2.49) | 8.67 (7.66) | 11.19 (11.33) |
| IL‐10 (pg/ml) | 0.25 (0.18) | 0.19 (0.17) | 0.29 (0.14) | 0.19 (0.08) | 0.29 (0.24) | 0.24 (0.18) | 0.26 (0.17) | 0.27 (0.29) | 0.15 (0.06) | 0.35 (0.67) | 0.25 (0.17) | 0.26 (0.29) | 0.22 (0.14) | 0.33 (0.25) |
| IL‐1 β (pg/ml) | 0.04 (0.04) | 0.05 (0.02) | 0.04 (0.04) | 0.03 (0.04) | 0.05 (0.04) | 0.05 (0.05) | 0.04 (0.04) | 0.06 (0.12) | 0.03 (0.02) | 0.08 (0.11) | 0.03 (0.02) | 0.07 (0.15) | 0.08 (0.19) | 0.06 (0.05) |
| IL‐6 (pg/ml) | 1.36 (3.68) | 0.26 (0.09) | 0.50 (0.39) | 0.65 (0.53) | 0.82 (0.68) | 0.96 (0.91) | 2.64 (6.32) | 3.36 (21.40) | 0.34 (0.20) | 5.26 (15.81) | 0.48 (0.28) | 1.16 (2.48) | 9.02 (41.81) | 1.19 (1.92) |
| IL‐8 (pg/ml) | 11.61 (6.10) | 9.17 (4.30) | 9.21 (2.60) | 10.64 (2.95) | 12.76 (11.19) | 12.82 (4.50) | 11.16 (4.18) | 16.09 (32.65) | 8.39 (3.51) | 9.52 (3.11) | 9.53 (3.44) | 12.78 (5.57) | 16.57 (13.02) | 25.19 (61.67) |
| IL‐1RA (pg/ml) | 287.27 (143.90) | 237.73 (70.53) | 266.35 (157.53) | 257.64 (102.38) | 285.26 (151.99) | 284.05 (127.07) | 311.91 (166.35) | 391.23 (447.15) | 278.84 (130.45) | 618.15 (1211.17) | 281.19 (147.41) | 466.07 (435.67) | 360.49 (282.33) | 348.91 (164.04) |
| sTNF‐RI (pg/ml) | 3978.10 (4064.95) | 2618.31 (446.61) | 2986.90 (444.16) | 2858.93 (495.42) | 3427.62 (1031.37) | 3597.62 (670.13) | 5416.41 (6992.57) a | 3515.85 (1160.92) | 2806.38 (505.08) | 3301.53 (1624.06) | 2892.08 (592.61) | 3495.22 (1215.66) | 3580.37 (820.30) | 3962.27 (1328.95) |
| sTNF‐RII (ng/ml) | 6149.07 (3913.30) | 4112.20 (683.75) | 5258.14 (793.49) | 4432.76 (1195.63) | 5481.48 (1710.10) | 5884.12 (1400.67) | 7775.34 (6386.86) | 5840.13 (2471.44) | 4758.99 (992.55) | 5077.89 (2166.04) | 4808.78 (888.46) | 5702.89 (2613.34) | 5699.14 (1926.31) | 7019.35 (3132.07) a |
Abbreviations: mg/ml, microgram per milliliter; N, number; ng/ml, nanogram per milliliter; pg/ml, picogram per milliliter.
ANOVA test, p < 0.05.
3.4. Cytokines by BMI and Smoking Status
Those who smoked (n = 24, 9.1%) reported higher TGF‐alpha levels (F = 4.19, p = 0.02) compared to those who never smoked (n = 126, 47.5%) or used to smoke (n = 115, 43.4%), see Table 3. Those with the highest weight (obese, n = 43, 16.2%), reported the highest levels of IL‐1RA (F = 7.98, p < 0.01), sTNFRI (F = 7.08, p < 0.01) and sTNFRII (F = 7.12, p < 0.01), followed by respondents who are overweight (n = 111, 41.9%), and those with normal weight (n = 111, 41.9%).
TABLE 3.
Biomarker levels stratified by smoking status, BMI, and number of self‐reported health conditions in means and standard deviations.
| Smoking status | BMI | Number of self‐reported health conditions | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Never (n = 126) | Former (n = 115) | Current (n = 24) | Normal (n = 111) | Overweight (n = 111) | Obese (n = 43) | No health conditions (n = 72) | 1 health conditions (n = 59) | ≥ 2 health conditions (n = 134) | |
| Mean age (SD) | 55.76 (15.51) | 63.00 (13.10) | 56.96 (16.57) b | 58.9 (16.11) | 59.09 (1.38) | 59.02 (13.19) | 51.71 (16.69) | 56.85 (15.46) | 63.89 (11.71) b |
| Men | 60 (47.6) | 57 (49.6) | 10 (41.7) | 49 (44.1) | 55 (49.5) | 23 (53.5) | 36 (50.0) | 33 (55.9) | 58 (43.3) |
| CRP (mg/ml) | 3.42 (5.31) | 4.22 (8.01) | 2.43 (2.59) | 2.98 (5.30) | 3.91 (7.96) | 4.85 (4.64) | 3.00 (5.59) | 3.15 (4.90) | 4.28 (7.46) |
| TGF‐α (pg/ml) | 11.06 (5.66) | 11.87 (6.18) | 15.02 a (8.03) | 11.36 (6.21) | 11.98 (6.49) | 12.33 (5.55) | 10.82 (4.33) | 12.25 (6.60) | 12.09 (6.86) |
| IL‐22 (pg/ml) | 1.45 (7.68) | 1.14 (5.23) | 2.31 (5.62) | 1.30 (7.95) | 1.12 (3.33) | 2.30 (8.37) | 2.22 (10.34) | 0.66 (1.17) | 1.27 (5.11) |
| IL‐17A (pg/ml) | 0.99 (1.43) | 0.96 (0.80) | 0.67 (0.59) | 0.79 (0.63) | 1.06 (1.04) | 1.09 (2.00) | 0.80 (0.69) | 0.94 (0.68) | 1.04 (1.43) |
| IL‐1 α (pg/ml) | 0.13 (0.71) | 0.10 (0.36) | 0.09 (0.19) | 0.08 (0.36) | 0.16 (0.76) | 0.08 (0.18) | 0.07 (0.13) | 0.19 (0.93) | 0.10 (0.45) |
| IFN‐γ (pg/ml) | 9.52 (19.42) | 8.80 (14.36) | 5.22 (3.30) | 7.67 (8.05) | 8.83 (20.23) | 11.71 (20.88) | 11.03 (24.95) | 6.31 (5.44) | 8.73 (13.56) |
| IL‐10 (pg/ml) | 0.28 (0.31) | 0.24 (0.14) | 0.25 (0.23) | 0.26 (0.18) | 0.27 (0.30) | 0.25 (0.19) | 0.26 (0.29) | 0.25 (0.22) | 0.27 (0.22) |
| IL‐1 β (pg/ml) | 0.05 (0.10) | 0.06 (0.09) | 0.05 (0.04) | 0.05 (0.09) | 0.05 (0.11) | 0.06 (0.05) | 0.05 (0.05) | 0.04 (0.04) | 0.06 (0.12) |
| IL‐6 (pg/ml) | 3.69 (22.55) | 1.10 (1.62) | 1.89 (4.69) | 0.90 (1.52) | 3.63 (23.41) | 3.06 (8.90) | 0.74 (0.76) | 1.82 (7.19) | 3.55 (21.47) |
| IL‐8 (pg/ml) | 15.57 (34.38) | 12.56 (5.29) | 11.98 (4.89) | 15.23 (35.72) | 13.67 (9.88) | 11.28 (4.11) | 15.73 (44.11) | 12.27 (7.79) | 13.69 (8.22) |
| IL‐1RA (pg/ml) | 354.20 (438.28) | 318.82 (211.41) | 380.92 (253.89) | 274.89 (192.44) | 340.27 (449.79) | 513.34 b (242.27) | 327.79 (512.94) | 300.81 (147.26) | 366.41 (280.44) |
| sTNF‐RI (pg/ml) | 3521.25 (1085.65) | 4017.38 (4269.07) | 3530.67 (1313.84) | 3371.09 (1011.15) | 3519.77 (961.03) | 5241.29 b (6812.23) | 3114.24 (690.07) | 3581.66 (1116.63) | 4145.47 (4009.19) |
| sTNF‐RII (ng/ml) | 5841.37 (2317.99) | 6198.31 (4174.77) | 5752.25 (2129.37) | 5532.23 (1772.03) | 5803.95 (1999.39) | 7633.64 b (6621.53) | 4955.74 (1130.06) | 5944.37 (2297.55) | 6567.67 (4126.94) |
Note: BMI was calculated as weight in kilograms divided by their squared height in meters and categorized into healthy weight (18–24.9 kg/m2), overweight (25–29.9 kg/m2), or obese (≥ 30 kg/m2). Alcohol use was categorized into occasional (0–14 g/day), moderate (14–41.9 g/day), or heavy (≥ 42 g/day) drinkers [22].
Abbreviations: mg/ml, microgram per milliliter; N, number; ng/ml, nanogram per milliliter; pg/ml, picogram per milliliter.
ANOVA test with p < 0.05.
ANOVA test with p < 0.01.
3.5. Cytokines by Self‐Reported Health Conditions
As expected, older participants reported more health conditions (F = 18.44, p < 0.01), more specifically heart disease (F = 20.81, p < 0.01), hypertension (F = 26.38, p < 0.01), and osteoarthritis (F = 26.02, p < 0.01), whereas depression was more common with younger participants (F = 7.34, p < 0.01), Tables 3 and 4. Diabetes (F = 5.31, p = 0.02) and osteoarthritis (F = 6.84, p < 0.01) were more common among women than among men.
TABLE 4.
Biomarker levels stratified by self‐reported health conditions in means and standard deviations.
| Self‐reported health condition in past 12 months | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Heart disease | Hypertension | Asthma/COPD | Diabetes | Depression | Osteoartritis | Rheumatoid arthritis | ||||||||
| No (n = 237) | Yes (n = 28) | No (n = 192) | Yes (n = 73) | No (n = 242) | Yes (n = 23) | No (n = 246) | Yes (n = 19) | No (n = 244) | Yes (n = 21) | No (n = 200) | Yes (n = 65) | No (n = 249) | Yes (n = 16) | |
| Mean age (SD) | 57.62 (15.08) | 70.79 (6.64) b | 56.23 (15.66) | 66.33 (9.83) b | 58.69 (15.05) | 62.43 (13.99) | 58.53 (15.13) | 65.21 (11.39) | 59.73 (15.10) | 50.62 (10.48) b | 56.46 (15.75) | 66.88 (8.37) b | 58.70 (15.15) | 63.81 (11.23) |
| Men | 112 (47.3) | 15 (11.8) | 90 (46.9) | 37 (50.7) | 116 (47.9) | 11 (47.8) | 118 (48.0) | 9 (47.4) a | 122 (50.0) | 5 (23.8) | 105 (52.5) | 22 (33.8) a | 123 (49.4) | 4 (25.0) |
| CRP (mg/ml) | 3.73 (6.68) | 3.22 (4.49) | 3.81 (7.21) | 3.34 (4.02) | 3.49 (6.36) | 5.69 (7.45) | 3.57 (6.43) | 5.08 (7.15) | 3.49 (5.32) | 5.88 (14.28) | 3.17 (5.07) | 5.28 (9.52) a | 3.67 (6.53) | 3.75 (5.82) |
| TGF‐α (pg/ml) | 11.99 (6.35) | 9.84 (4.42) | 12.05 (6.41) | 11.06 (5.61) | 11.58 (6.14) | 14.02 (6.76) | 11.57 (6.09) | 14.53 (7.33) | 11.56 (5.99) | 14.23 (8.09) | 11.48 (5.83) | 12.68 (7.23) | 11.95 (6.32) | 9.16 (3.34) |
| IL‐22 (pg/ml) | 1.50 (6.87) | 0.42 (0.44) | 1.31 (6.43) | 1.61 (6.78) | 1.41 (6.79) | 1.15 (1.78) | 1.41 (6.72) | 1.13 (2.20) | 1.47 (6.78) | 0.44 (0.56) | 1.27 (6.28) | 1.75 (7.24) | 1.44 (6.72) | 0.57 (0.57) |
| IL‐17A (pg/ml) | 0.95 (1.17) | 0.99 (0.73) | 0.96 (1.22) | 0.93 (0.86) | 0.92 (1.11) | 1.32 (1.29) | 0.95 (1.15) | 0.95 (0.91) | 0.96 (1.16) | 0.87 (0.79) | 0.98 (1.21) | 0.87 (0.82) | 0.93 (1.10) | 1.34 (1.53) |
| IL‐1 α (pg/ml) | 0.12 (0.58) | 0.06 (0.12) | 0.12 (0.59) | 0.10 (0.43) | 0.12 (0.57) | 0.04 (0.05) | 0.12 (0.57) | 0.04 (0.06) | 0.08 (0.26) | 0.54 (1.70) b | 0.12 (0.57) | 0.09 (0.46) | 0.12 (0.56) | 0.04 (0.04) |
| IFN‐γ (pg/ml) | 8.94 (17.29) | 7.74 (4.56) | 8.88 (18.20) | 8.65 (10.52) | 8.41 (14.98) | 13.05 (27.44) | 8.91 (17.00) | 7.65 (4.57) | 9.20 (17.04) | 4.39 (3.22) | 9.45 (18.52) | 6.82 (6.06) | 8.91 (16.90) | 7.34 (4.28) |
| IL‐10 (pg/ml) | 0.26 (0.25) | 0.28 (0.20) | 0.25 (0.24) | 0.28 (0.25) | 0.26 (0.23) | 0.31 (0.37) | 0.26 (0.25) | 0.27 (0.12) | 0.27 (0.25) | 0.19 (0.11) | 0.27 (0.27) | 0.23 (0.13) | 0.26 (0.24) | 0.30 (0.18) |
| IL‐1 β (pg/ml) | 0.05 (0.08) | 0.08 (0.16) | 0.06 (0.09) | 0.05 (0.10) | 0.05 (0.10) | 0.07 (0.06) | 0.05 (0.10) | 0.07 (0.06) | 0.06 (0.10) | 0.04 (0.04) | 0.05 (0.05) | 0.08 (0.17) b | 0.05 (0.08) | 0.10 (0.20) a |
| IL‐6 (pg/ml) | 2.42 (16.45) | 2.18 (4.91) | 2.56 (18.11) | 1.97 (5.22) | 2.52 (16.36) | 1.14 (1.16) | 2.09 (15.73) | 6.33 (14.11) | 2.53 (16.29) | 0.80 (0.87) | 1.24 (4.27) | 6.00 (30.75) a | 2.34 (16.02) | 3.27 (7.48) |
| IL‐8 (pg/ml) | 14.01 (25.30) | 13.25 (4.28) | 13.93 (27.57) | 13.95 (9.58) | 13.83 (24.97) | 15.00 (7.83) | 13.87 (24.83) | 14.79 (5.46) | 14.23 (24.93) | 10.53 (3.78) | 13.66 (26.95) | 14.80 (10.03) | 13.57 (24.32) | 19.55 (16.96) |
| IL‐1RA (pg/ml) | 344.10 (354.03) | 316.89 (196.37) | 328.54 (358.83) | 374.39 (287.19) | 331.76 (335.17) | 440.28 (387.58) | 327.46 (334.93) | 518.66 (372.46) a | 341.36 (347.84) | 339.54 (248.32) | 333.13 (362.21) | 366.50 (262.84) | 343.60 (346.58) | 304.31 (233.98) |
| sTNF‐RI (pg/ml) | 3750.52 (3094.12) | 3634.58 (1040.93) | 3688.04 (3366.14) | 3869.51 (1316.40) | 3749.28 (3069.38) | 3622.32 (904.15) | 3517.64 (2963.71) | 3738.22 (2943.95) a | 3772.90 (3058.83) | 3336.90 (756.57) | 3701.99 (3298.74) | 3851.44 (1338.35) | 3735.19 (3003.34) | 3785.27 (1850.79) |
| sTNF‐RII (ng/ml) | 5975.91 (3369.14) | 6096.99 (1911.53) | 5785.26 (3389.78) | 6521.19 (2775.66) | 5965.90 (3285.88) | 6228.22 (2810.77) | 5831.85 (3131.56) | 8012.00 (4013.14) a | 6041.64 (3338.99) | 5376.76 (1731.64) | 5905.99 (3495.61) | 6247.38 (2286.09) | 5949.33 (3240.08) | 6599.86 (3337.98) |
Abbreviations: mg/ml, microgram per milliliter; N, number; ng/ml, nanogram per milliliter; pg/ml, picogram per milliliter.
ANOVA test with p < 0.05.
ANOVA test with p < 0.01.
sTNFRII levels increased as the number of self‐reported health conditions increased (F = 6.00, p < 0.01). Those who were diagnosed with diabetes (n = 19, 7.2%) reported higher levels of IL‐1RA (F = 5.65, p = 0.02), sTNFR‐I (F = 5.82, p = 0.02), and sTNFR‐II (F = 8.19, p < 0.01) compared to those without diabetes. Individuals with osteoarthritis showed higher levels of CRP (F = 5.23, p = 0.02), IL‐1beta (F = 8.08, p < 0.01), and IL‐6 (F = 4.54, p = 0.03) than those without.
4. Discussion
This study provides reference distribution data on 12 inflammation‐related serum biomarkers derived from a Dutch population‐based sample of adults without a history of cancer. These distributions should be interpreted within the context of the Dutch general population, the applied sampling strategy, and the standardized laboratory procedures used in this study. The availability of such datasets is essential for interpreting immune profiles in the context of cancer survivorship, yet should not be interpreted as clinical reference thresholds.
Cancer survivors often display persistently elevated levels of inflammatory markers, even long after the completion of treatment, a phenomenon that cannot be fully attributed to cancer or its therapies alone. Enabling reference distributions of a healthy, age‐ and sex‐matched population is therefore critical for distinguishing cancer‐related biological alterations from changes associated with normal aging, lifestyle factors, or comorbid health conditions. The data presented herein lay a necessary foundation for exploring the biological underpinnings of PROs such as fatigue, pain, depressive symptoms, and diminished QoL among cancer patients, while accounting for non‐cancer‐related influences on systemic inflammation.
Our findings align with previous population‐based studies demonstrating that cytokine concentrations vary significantly by demographic and health‐related factors.
Given the descriptive aim of establishing reference distributions, the comparative analyses in this study were intended to support interpretation of distributional patterns rather than provide confirmatory evidence of biological differences between groups. Nevertheless, older individuals, smokers, and those with higher BMI or multiple comorbid conditions exhibited elevated levels of pro‐inflammatory cytokines including IL‐1RA, IL‐6, CRP, sTNFRI, and sTNFRII. The observation that IL‐6 and IL‐1β levels are higher among individuals with osteoarthritis corroborates previous studies linking musculoskeletal disorders to systemic inflammation [24]. Additionally, the elevated sTNFRII and IL‐1RA levels in participants with diabetes mirror findings from studies on metabolic inflammation [25, 26]. However, while our findings did not uniformly support the “inflammaging” hypothesis as broadly described in the gerontological literature [27], they did align with previous evidence indicating that advancing age is consistently associated with elevated levels of specific inflammatory markers—most notably soluble TNF receptors I and II (sTNFRI and sTNFRII) [28].
While international initiatives such as the SHIP‐TREND cohort in Germany [29] and the cytokine dataset reported by Li et al. [30] have contributed valuable normative data, differences in methodology, population structure, and assay systems limit direct comparability. Our study adds to this literature by using a harmonized protocol (identical to that used in cancer cohorts), enabling more precise comparisons across populations while accounting for demographic differences and the presence of common health conditions.
A key strength of this study is the stratification of cytokine levels according to lifestyle factors such as smoking status, BMI, and prevalent health conditions, resulting in nuanced and clinically relevant reference values. This approach enables more accurate differentiation between biomarker–PRO associations that are specific to oncological populations and those that also manifest in the general population. Moreover, it facilitates meaningful comparisons with inflammatory profiles observed in other chronic disease contexts, including cardiometabolic, rheumatologic, and neurodegenerative disorders. The use of a centralized biobank and standardized laboratory protocols further strengthens the methodological rigor and reliability of our findings.
While cytokine concentrations may show some variability across analytical platforms, such differences are a well‐recognized aspect of inflammatory biomarker research. The reference distributions presented here were obtained under standardized pre‐analytical and laboratory procedures, providing a consistent framework for interpreting inflammatory profiles in clinical and survivorship studies. These data may therefore support meaningful comparative research, while acknowledging that methodological factors can contribute to variation in absolute concentrations across studies. Overall, comparisons across cohorts and settings remain informative when interpreted considering assay‐related context.
Several limitations warrant mentioning. First, given the number of biomarkers, stratifications, and subgroup analyses, the risk of type I error is increased. Findings should therefore be interpreted as exploratory rather than confirmatory. Additionally, these reference distributions should primarily be interpreted within the context of Dutch population‐based research settings using comparable recruitment procedures and laboratory methodologies. Generalization to other populations, ethnic backgrounds, healthcare systems, or assay platforms should therefore be undertaken cautiously. Subsequently, cytokine concentrations are known to vary across analytical platforms, assay manufacturers, sample matrices, and pre‐analytical handling procedures. Although all samples in the present study were processed using harmonized protocols and measured on a single standardized MSD platform, absolute cytokine concentrations may differ from those reported in studies using alternative assay systems. Therefore, comparisons across studies should be interpreted within the context of methodological differences. The reported means and standard deviations are intended to describe variation within a population‐based sample and should not be interpreted as diagnostic thresholds or normative cut‐offs for clinical use. Moreover, as is common in cytokine research, several biomarkers included non‐detectable values that required imputation. Although substitution methods are commonly used in cytokine research, differences in handling non‐detectable values should be considered when comparing results across studies. As a sensitivity analysis, descriptive statistics were compared between the imputed and non‐imputed datasets. Biomarker means and standard deviations were highly similar across both approaches. Furthermore, although the panel is representative of the general Dutch population, participants willing to provide blood samples were on average younger and somewhat healthier than those who did not. Therefore, the presented values should be interpreted as reference distributions derived from a relatively healthy volunteer sample within a population‐based cohort, which could result in lower observed biomarker levels. This potential healthy volunteer effect should be considered when interpreting these distributions as a reference framework.
While the prevalence levels of self‐reported health conditions and BMI distribution were largely similar to the Dutch adult population [31, 32], the number of smokers in our sample was lower [33], suggesting some degree of health‐related self‐selection. Consequently, caution is warranted when generalizing these values beyond similar research settings. In addition, while the sample size was adequate for general comparisons, stratification by specific conditions (e.g., diabetes or depression) resulted in small subgroups, reducing the stability and precision of subgroup‐specific estimates. Hence this limits statistical precision and requires careful interpretation of subgroup‐specific estimates.
In conclusion, our study provides contextual reference distributions for key inflammatory biomarker data stratified by sex, age, lifestyle factors, and common health conditions within a Dutch population‐based cohort. These data offer a valuable contextual framework for interpreting biomarker‐PRO associations in both (cancer) patients and the general population, while recognizing that inflammatory marker levels may also reflect normal aging, lifestyle, comorbid health conditions, and methodological context. To facilitate future research, all raw biomarker data used in this manuscript and all questionnaire data on relevant PROs (including depression, anxiety, fatigue, and QoL) are available for non‐commercial scientific research upon request, subject to study question, privacy, and confidentiality restrictions, via www.profilesregistry.nl.
Author Contributions
Dounya Schoormans: conceptualization (equal), data curation (equal), formal analysis (equal), funding acquisition (equal), investigation (equal), methodology (equal), project administration (equal), supervision (equal), validation (equal), visualization (equal), writing – original draft (equal), writing – review and editing (equal). Nicole Horevoorts: conceptualization (equal), data curation (equal), funding acquisition (equal), project administration (equal), writing – review and editing (equal). Marije Oudejans: conceptualization (equal), data curation (equal), methodology (equal), project administration (equal), writing – review and editing (equal). Marjo van de Waarenburg: data curation (equal), methodology (equal), project administration (equal), validation (equal), writing – review and editing (equal). Floortje Mols: conceptualization (equal), data curation (equal), funding acquisition (equal), investigation (equal), methodology (equal), project administration (equal), supervision (equal), writing – review and editing (equal).
Funding
The present research was supported by the Netherlands Comprehensive Cancer Organization, Utrecht, the Netherlands; the Center of Research on Psychological disorders and Somatic diseases (CoRPS), Tilburg University, the Netherlands; and an Investment Grant Large (2016/04981/ZONMW‐91101002) of the Dutch Research Council (The Hague, The Netherlands). These funders had no role in the design and conduct of the study, preparation of the manuscript, and decision to submit the manuscript for publication.
Disclosure
The authors have nothing to report.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Data S1: Supporting Information.
Acknowledgements
We would like to thank all participants for their (ongoing) participation. Also, we would like to thank Centerdata for their continuing efforts to further develop the PROFILES registry and maintain its websites and panel management system.
Data Availability Statement
PROFILES‐data and stored blood samples are freely available upon request for scientific purposes (www.profilesregistry.nl).
References
- 1. International_agency_for_resesarch_on_cancer , “Globocan,” (2020), https://gco.iarc.fr/today/data/factsheets/populations/900‐world‐fact‐sheets.pdf.
- 2. Shapiro C. L., “Cancer Survivorship,” New England Journal of Medicine 379, no. 25 (2018): 2438–2450. [DOI] [PubMed] [Google Scholar]
- 3. Mols F., Schoormans D., Oerlemans S., et al., “Patient‐Reported Outcomes in Cancer Survivorship: Insights From Two Decades of Population‐Based PROFILES Registry Research,” Journal of Cancer Survivorship 20 (2024): 827–838. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Sprangers M. A., Sloan J. A., Barsevick A., et al., “Scientific Imperatives, Clinical Implications, and Theoretical Underpinnings for the Investigation of the Relationship Between Genetic Variables and Patient‐Reported Quality‐Of‐Life Outcomes,” Quality of Life Research 19, no. 10 (2010): 1395–1403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Barsevick A., Frost M., Zwinderman A., Hall P., and Halyard M., “I'm So Tired: Biological and Genetic Mechanisms of Cancer‐Related Fatigue,” Quality of Life Research 19, no. 10 (2010): 1419–1427. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. van de Poll‐Franse L. V., Horevoorts N., Schoormans D., et al., “Measuring Clinical, Biological, and Behavioral Variables to Elucidate Trajectories of Patient‐Reported Outcomes: The PROFILES Registry,” JNCI Journal of the National Cancer Institute 114, no. 6 (2022): 800–807. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Kim H. J., Barsevick A. M., Fang C. Y., and Miaskowski C., “Common Biological Pathways Underlying the Psychoneurological Symptom Cluster in Cancer Patients,” Cancer Nursing 35, no. 6 (2012): E1–E20. [DOI] [PubMed] [Google Scholar]
- 8. Miller A. H., Ancoli‐Israel S., Bower J. E., Capuron L., and Irwin M. R., “Neuroendocrine‐Immune Mechanisms of Behavioral Comorbidities in Patients With Cancer,” Journal of Clinical Oncology 26, no. 6 (2008): 971–982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Bower J. E., “Cancer‐Related Fatigue—Mechanisms, Risk Factors, and Treatments,” Nature Reviews. Clinical Oncology 11, no. 10 (2014): 597–609. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Manigault A. W., Ganz P. A., Irwin M. R., Cole S. W., Kuhlman K. R., and Bower J. E., “Moderators of Inflammation‐Related Depression: A Prospective Study of Breast Cancer Survivors,” Translational Psychiatry 11, no. 1 (2021): 615. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Teissier T., Boulanger E., and Cox L. S., “Interconnections Between Inflammageing and Immunosenescence During Ageing,” Cells 11, no. 3 (2022): 359. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Martinez de Toda I., González‐Sánchez M., Díaz‐Del Cerro E., Valera G., Carracedo J., and Guerra‐Pérez N., “Sex Differences in Markers of Oxidation and Inflammation. Implications for Ageing,” Mechanisms of Ageing and Development 211 (2023): 111797. [DOI] [PubMed] [Google Scholar]
- 13. Liqiang S., Fang‐Hui L., Minghui Q., and Haichun C., “Threshold Effect and Sex Characteristics of the Relationship Between Chronic Inflammation and BMI,” BMC Endocrine Disorders 23, no. 1 (2023): 175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Lee J., Taneja V., and Vassallo R., “Cigarette Smoking and Inflammation: Cellular and Molecular Mechanisms,” Journal of Dental Research 91, no. 2 (2012): 142–149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Laveti D., Kumar M., Hemalatha R., et al., “Anti‐Inflammatory Treatments for Chronic Diseases: A Review,” Inflammation & Allergy Drug Targets 12, no. 5 (2013): 349–361. [DOI] [PubMed] [Google Scholar]
- 16. Scherpenzeel A. and Das M., “True Longitudinal and Probability‐Based Internet Panels: Evidence From The Netherlands,” in Social and Behavioral Research and the Internet: Advances in Applied Methods and Research Strategies, ed. Das P. E. N. and Kaczmirek L. (Taylor & Francis, 2011), 77–104. [Google Scholar]
- 17. van de Poll‐Franse L. V., Horevoorts N., van Eenbergen M., et al., “The Patient Reported Outcomes Following Initial Treatment and Long Term Evaluation of Survivorship Registry: Scope, Rationale and Design of an Infrastructure for the Study of Physical and Psychosocial Outcomes in Cancer Survivorship Cohorts,” European Journal of Cancer 47, no. 14 (2011): 2188–2194. [DOI] [PubMed] [Google Scholar]
- 18. Mols F., van Cappellen‐van Maldegem S., Hoedjes M., et al., “Remote Blood Collection Among Cancer Patients and Age‐ and Sex Matched Controls for Biomarker and Genetic Analyses Using the PROFILES Registry,” Support Care Cancer 34, no. 61 (2025): 61. [DOI] [PubMed] [Google Scholar]
- 19. Mols F., Schoormans D., Netea‐Maier R., et al., “Determinants and Mediating Mechanisms of Quality of Life and Disease‐Specific Symptoms Among Thyroid Cancer Patients: The Design of the WaTCh Study,” Thyroid Research 16, no. 1 (2023): 23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. van Cappellen‐van Maldegem S. J. M., Mols F., Horevoorts N., et al., “Towards OPtimal TIming and Method for Promoting sUstained Adherence to Lifestyle and Body Weight Recommendations in postMenopausal Breast Cancer Survivors (The OPTIMUM‐Study): Protocol for a Longitudinal Mixed‐Method Study,” BMC Womens Health 21, no. 1 (2021): 268. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. MesoScaleDiscovery , “Rockville, MD: Meso Scale Diagnostics”.
- 22. Choi N. G. and Dinitto D. M., “Heavy/Binge Drinking and Depressive Symptoms in Older Adults: Gender Differences,” International Journal of Geriatric Psychiatry 26, no. 8 (2011): 860–868. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Sangha O., Stucki G., Liang M. H., Fossel A. H., and Katz J. N., “The Self‐Administered Comorbidity Questionnaire: A New Method to Assess Comorbidity for Clinical and Health Services Research,” Arthritis and Rheumatism 49, no. 2 (2003): 156–163. [DOI] [PubMed] [Google Scholar]
- 24. Gallo J., Raska M., Kriegova E., and Goodman S. B., “Inflammation and Its Resolution and the Musculoskeletal System,” Journal of Orthopaedic Translation 10 (2017): 52–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Herder C., Brunner E. J., Rathmann W., et al., “Elevated Levels of the Anti‐Inflammatory Interleukin‐1 Receptor Antagonist Precede the Onset of Type 2 Diabetes: The Whitehall II Study,” Diabetes Care 32, no. 3 (2009): 421–423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Zhao L., Hu H., Zhang L., et al., “Inflammation in Diabetes Complications: Molecular Mechanisms and Therapeutic Interventions,” MedComm 5, no. 4 (2020): e516. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Franceschi C., Capri M., Monti D., et al., “Inflammaging and Anti‐Inflammaging: A Systemic Perspective on Aging and Longevity Emerged From Studies in Humans,” Mechanisms of Ageing and Development 128, no. 1 (2007): 92–105. [DOI] [PubMed] [Google Scholar]
- 28. Parker D., Sloane R., Pieper C. F., et al., “Age‐Related Adverse Inflammatory and Metabolic Changes Begin Early in Adulthood,” Journals of Gerontology. Series A, Biological Sciences and Medical Sciences 74, no. 3 (2019): 283–289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Ameling S., van der Auwera S., Holtfreter S., et al., “Cytokine Atlas of the Population‐Based Cohort SHIP‐TREND‐0 ‐ Associations With Age, Sex, and BMI,” Cytokine 189 (2025): 156896. [DOI] [PubMed] [Google Scholar]
- 30. Li Y., Yi J. S., Russo M. A., Rosa‐Bray M., Weinhold K. J., and Guptill J. T., “Normative Dataset for Plasma Cytokines in Healthy Human Adults,” Data in Brief 35 (2021): 106857. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. VZinfo , “Informatie over volksgezondheid en zorg,” (2025), https://www.vzinfo.nl/.
- 32. CBS , “Statistics,” https://opendata.cbs.nl/statline/#/CBS/nl/.
- 33. Trimbos‐instituut , “CBS en RIVM,” (2024), https://www.trimbos.nl/wp‐content/uploads/2025/02/TRI64‐010‐Factsheet‐Kerncijfers‐roken‐2023.pdf.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1: Supporting Information.
Data Availability Statement
PROFILES‐data and stored blood samples are freely available upon request for scientific purposes (www.profilesregistry.nl).