Skip to main content
NCBI home page
BMC Oral Health logoLink to BMC Oral Health
. 2025 Nov 22;25:1902. doi: 10.1186/s12903-025-07368-2

Methodological comparison between salivary and plasma inflammatory biomarkers in third molar surgery patients

Lars B Eriksson 1,2,8,✉, Anders Larsson 3, Mats Eriksson 1,4, Åke Tegelberg 1,5, Andreas Thor 1,6, Torsten Gordh 1,7
PMCID: PMC12703913  PMID: 41272580

Abstract

Background

Changes in saliva biomarkers levels may be an inflammatory response to local surgical trauma or other threats. The aim of the study was to compare saliva and plasma regarding the expression of a large set of inflammatory biomarkers to find clinically useful biomarkers in saliva.

Methods

Both saliva and blood samples were collected from 165 individuals. For every patient, the samples were collected on the same occasion. Saliva and plasma protein levels were analysed using the OLINK Proseek inflammation panel measuring 92 cytokines, chemokines, and growth factors (CCGF). The levels of individual CCGF were compared between saliva and plasma. A Spearman rank test was used to find correlations (rs).

Results

Of the 92 inflammatory biomarkers, 71 were detected in plasma, 63 in saliva, and 58 in both saliva and plasma. IL-6 (rs = 0.1535, p = 0.048) and CST5 (rs = -0.2542, p = 0.00098) showed significant correlations between their expression levels in saliva and plasma. Furthermore, 36 significantly correlated heterogeneous cytokine pairs were identified. In only one pair was rs ≥ ± 0.25; in all other cases the correlations were even weaker. CCGF, including IL-8, VEGFA, CDCP1, IL-6, IL-1 alpha, OSM, TNFSF14, CCL28, EN-RAGE, and CASP-8, were expressed much more strongly in saliva than in plasma.

Conclusion

We found major differences in the levels of inflammatory biomarkers in saliva versus plasma when analysed with the OLINK method. A compound of the six most prominent proteins in saliva (IL-8, IL-1 alpha, IL-6, OSM, CST5 and CCL28) are expressed more strikingly in saliva than in plasma. They are also connected by certain inflammatory functions. Obviously, saliva samples do not give the same information on inflammation processes as those found in plasma. This information may be important for future inflammation studies.

Trial registration

EudraCT under number, 2014-004235-39 (29/09/2014). ClinicalTrials.gov, ID: NCT04459377 (15/07/2020).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12903-025-07368-2.

Keywords: Saliva, Plasma, Biomarkers, Cytokines, Inflammation, Inflammation mediators, Pain

Background

There are several studies measuring CCFG in peripheral fluids such as plasma, saliva, or urine, but only a few of them have directly compared proteins in saliva versus plasma from the same individuals and at the same time [1–5]. Saliva is easily accessible for sampling with a non-invasive method, in contrast to sampling of blood, cerebrospinal fluid, or even urine, especially when integrity is a concern. CCGF are of the utmost importance in the inflammation process and, thus, crucial for understanding inflammatory pain. The pathogenesis of acute and chronic pain is not fully understood. Through a better understanding of the inflammatory proteins and their associations across different compartments, it may be possible to gain more insight into the complex processes behind nociception and pain experience. The immune system partially acts as a diffuse sense organ, when activated, by signalling the brain about peripheral events [6]. Inflammation is a biological response of the immune system to harmful stimuli such as pathogens, damaged cells, or irritants [7]. In pathological manifestations in the oral cavity - such as in oral potentially malignant disorders including leucoplakia - salivary cytokines are increased; this is particularly evident in oral cancer cases [8, 9]. In one study from 2022, 10 cytokines in saliva vs. plasma were compared in 71 elderly participants, concluding that inflammatory biomarkers in saliva are associated with those found in plasma, which suggests that there is a similar inflammatory mechanism between saliva and plasma [3]. Third molar surgery is a well-established pain model frequently used in pain research especially in pharmacological trials [10–14]; normal findings in saliva and plasma in that population are useful to evaluate the model. To clarify if specific biomarkers quantified in saliva can substitute such quantification in plasma, further scientific comparisons are needed. To our knowledge it has not been made previously in a cohort like ours.

Patients and methods

The aim of this study was to compare salivary with plasma expression of inflammatory biomarkers using paired samples taken from patients, and to find correlations or proteins primarily expressed at high levels in saliva.

Patients referred to the Department of Oral and Maxillofacial Surgery at the Falu County Hospital, Sweden, for mandibular third molar removal, were considered for inclusion. The study focused on men and women from 18 to 44 years old, with a bodyweight of 50 to 120 kg. Patients who met the exclusion criteria (Table 1.) were not invited to participate in the study. Those who signed a written informed-consent form were then included in the study. All participants fasted from 12.00 pm on the night before the samples of saliva and plasma were collected. These patients took part in a clinical trial, reported elsewhere, but all sampling of data presented in this study was made before the patients were exposed to any intervention [15, 16]. The patients and surgical procedures have been described in detail elsewhere [15].

Table 1.

Exclusion criteria

Exclusion criteria
Medication: analgesics, hypnotics (the week before surgery), thyroid hormones, psychotropic drugs or MAO-inhibitors.
Hypertension [> 150/95 mmHg at screening], Congestive heart failure, Psychosis, Epilepsy, Hyperthyroidism, Myasthenia gravis, Glaucoma, Verified sleep apnoea, Diabetes (insulin dependent), Porphyria, Pregnancy, Breast-feeding, Blood infections (Hepatitis B, C and HIV), Inability to comprehend written or spoken information, known hypersensitivity or allergy to midazolam, ketamine, ibuprofen or local anaesthetic.
Open in a new tab

Saliva collection

The saliva samples were collected between 8 am and 2 pm by asking each patient to spit into a funnel, fitted in a plastic tube (Sarstedt® 3 ml), to reach at least 1 ml saliva (Fig. 1). Within one hour, the samples were frozen and stored at −80 °C until they were analysed.

Fig. 1.

Fig. 1

Open in a new tab

Plastic funnel and Sarstedt® 3 ml tube for saliva sampling

Plasma collection

Plasma samples were collected immediately after the saliva collection by placing an intravenous (IV) cannula (BD Venflon® Ø 1.3 mm) into a superficial vein on the back of the hand or on the forearm (Fig. 2). The blood was collected in EDTA tubes (BD Vacutainer®). The samples were prepared by centrifugation for 5 min at 2400 g, and the plasma portions were stored in aliquots at −80 °C within one hour from sampling until the analysis took place. Preparation time affects a majority of proteins measured in plasma and there is a risk of causing pre-analytic errors when the centrifugation is delayed, especially by more than 1 h [17, 18]. 165 patients were enrolled, resulting in 165 paired samples (n = 330) of plasma plus saliva to analyse. The time from collecting the samples to preparation was minimized and the time from collecting the samples to they were frozen and stored did not exceed 60 min. All saliva samples as was all the blood samples, were handled in the same way, to avoid preanalytical errors.

Fig. 2.

Fig. 2

Open in a new tab

Sampling of venous blood from superficial arm vein

Proximity extension assay (PEA)

Briefly, 1 µL plasma from every aliquot was used in the OLINK Target 96 Inflammation panel analysis (Olink Bioscience, Uppsala, Sweden). The procedure was the same for the saliva samples. The samples were analysed according to the manufacturer’s instructions, applying a proximity extended assay (PEA) technology, where 92 inflammatory-related proteins were measured [19, 20]. The plasma was mixed with 3 µL of incubation mix, which contained two probes (antibodies labelled with unique corresponding DNA oligonucleotides). The mixture was incubated at 8 °C overnight. Following this, 96 µL of extension mix, containing the PEA enzyme and PCR reagents, was added. The samples were then incubated for 5 min at room temperature before being transferred to a thermal cycler for 17 cycles of DNA amplification. A 96.96 Dynamic Array IFC (Fluidigm, South San Francisco, CA, USA) was prepared and primed according to the manufacturer’s instructions. In a separate plate, 2.8 µL of the sample mixture was combined with 7.2 µL of detection mix, from which 5 µL was loaded into the right side of the primed 96.96 Dynamic Array IFC. Unique primer pairs for each cytokine were loaded into the left side of the 96.96 Dynamic Array IFC, and the protein expression program was run on the Fluidigm Biomark reader according to the Proseek instructions. Each Proseek kit measured 92 inflammatory CCGF biomarkers in total (Supplementary Table 6). The quantification of these protein concentrations is expressed logarithmically as normalized protein expression (NPX) [21]. The NPX value is an arbitrary unit on a Log2 scale meaning a difference in one unit of NPX corresponds to a doubling of the concentration.

Statistics

All the output data from the OLINK set-up was utilized, analysing both saliva and plasma. 71 proteins in plasma, 63 proteins in saliva, and 58 proteins in both saliva and plasma out of 92 analysed showed >80% samples with NPX values above the limit of detection (LOD). LOD was defined as three standard deviations (SD) above the average of the negative controls. Spearman´s rank correlation coefficient was calculated using Statistica (StatSoft Version 14, Tulsa, OK, USA) to examine the correlation of expression of various cytokines (Table 4.) as NPX in saliva vs. plasma, (rs = ± 1) [22]. For some descriptive parts of the data, Jamovi (ver. 2.3.28) was used.

Table 4.

Inflammatory biomarkers with significant correlation between expression in saliva vs. plasma

Saliva cytokine Uniprot-ID Plasma cytokine Spearman(rs) p-value s/ns
IL6 P05231 IL6 0.153554 0.048937 S
CST5 P28325 IL-17 A 0.153931 0.048378 S
CST5 P28325 CXCL11 0.171429 0.027692 S
CST5 P28325 CXCL9 0.155734 0.045778 S
CST5 P28325 CST5 −0.254204 0.000985 S
OSM P13725 IL6 0.166711 0.032341 S
FGF-21 Q9NSA1 GDNF −0.186047 0.016731 S
FGF-21 Q9NSA1 CXCL9 0.163984 0.035321 S
IL10 P22301 AXIN1 0.154051 0.048201 S
FGF-19 O95750 IL-17 A −0.156349 0.044918 S
FGF-19 O95750 TRAIL 0.198042 0.010777 S
FGF-19 O95750 CST5 0.190500 0.014250 S
CST5 P28325 CCL11 0.186838 0.016264 S
FGF-21 Q9NSA1 OSM −0.160254 0.039767 S
Flt3L P49771 IL18 0.190564 0.014216 S
FGF-19 O95750 IL18 0.170855 0.028225 S
FGF-19 O95750 IL-15RA 0.191099 0.013942 S
FGF-19 O95750 IL-18R1 0.193329 0.012846 S
IL6 P05231 IL-24 0.190666 0.014164 S
FGF-21 Q9NSA1 ARTN −0.158114 0.042524 S
FGF-21 Q9NSA1 CXCL10 0.212342 0.006179 S
Flt3L P49771 CD5 0.156090 0.045279 S
FGF-19 O95750 DNER 0.195825 0.011711 S
IL6 P05231 CCL20 0.201867 0.009318 S
CST5 P28325 MCP-2 0.185895 0.016822 S
CST5 P28325 CCL20 0.199685 0.010127 S
OSM P13725 MCP-2 −0.171520 0.027608 S
OSM P13725 TNFB −0.184818 0.017478 S
TGF-alpha P01135 CCL25 0.177012 0.022938 S
TGF-alpha P01135 TNFRSF9 0.206492 0.007791 S
FGF-21 Q9NSA1 IFN-gamma 0.161259 0.038525 S
IL10 P22301 FGF-19 −0.207023 0.007630 S
IL10 P22301 TNFB −0.212160 0.006224 S
Flt3L P49771 TNFRSF9 0.159100 0.041235 S
FGF-19 O95750 TNFRSF9 0.164184 0.035094 S
FGF-19 O95750 ST1A1 0.163751 0.035585 S
Open in a new tab

Eight CCGF in plasma (see Table 3.) were found to have significant changes due to surgical trauma (Eriksson et al., submitted 2025), and were tested for correlations to 92 CCGF in saliva. Of the 92 CCGF, 36 pairs from 8 × 92 possible combinations were found to have a correlation between saliva and plasma. Valid N = 165. s/ns = significant/not significant

When translating the numeric value into descriptive words, authors have chosen different words. For example, the rs coefficient ± 0.2 or 0.3 has been described as weak, poor, moderate, or fair [23, 24]; in this study, we have chosen the word weak.

Ethics

The Swedish Ethical Review Authority revised and approved the study protocol (Reference number: 2015/378. Date: 02/12/2015).

Prior to any study procedure, we obtained informed consent from all patients. Good Clinical Practice (GCP), General Data Protection Regulation (GDPR;2016/679), and the Helsinki Declaration were followed throughout the study.

Results

One hundred and sixty-five patients, between 18 and 44 years old and weighing between 50 and 111 kg were included in this study (Table 2).

Table 2.

Basic characteristics of the population (N = 165)

Valid N Mean Median Lower
Quartile		 Upper
Quartile
Sex 165

112 females

53 males
Age year 165 29.1 29 23 35
Weight kg 165 74.4 72.9 64.9 82.1
BMI kg/m2 165 25.3 24.3 22.3 27.7
Hb g/L 165 137 134 129 145
EVF 165 0.41 0.41 0.39 0.43
WBC x109/L 165 5.92 5.5 4.6 6.7
Plt x109/L 165 244 239 207 273
Alb g/L 165 41.7 42 40 44
Crea µmol/L 165 68.1 67 59 76
Open in a new tab

EVF Erytrocyte volume fraction, WBC White blood cell count, Plt Platelet count, Alb Albumin, Crea Creatinine

IL-6 and CST5 were significantly correlated to the same protein in saliva (Table 3.) (Fig. 3). Out of eight proteins in plasma, significantly altered by the surgical trauma of third molar surgery, only two correlated to the expression of the same protein in saliva (Eriksson et al. submitted 2025) (seen in Table 3.). Thirty-six correlated protein pairs between saliva and plasma were found when testing the eight saliva proteins (IL6, CST5, OSM, TGF-alpha, FGF-21, IL10, Flt3L, and FGF-19) against 92 different proteins in plasma. In only one pair the correlation was rs ≥ ± 0.25, and in only six pairs the correlation was rs ≥ ± 0.20. The p-value in all these cases was < 0.05. (Table 4.). When comparing median NPX-value for each protein in saliva vs. plasma (Fig. 4), four proteins uPA, CD40, NRTN, and CSF-1, were found to have a less than 5% difference in relative concentration between saliva and plasma (Table 5). When these four were tested for correlations using the Spearman rank scale, only weak correlations, rs=0.151 or less, were found (Table 5.). Some of the 92 CCGF were more strongly expressed in saliva than in plasma: IL-8, VEGFA, CDCP1, IL-6, IL-1 alpha, OSM, TNFSF14, CCL28, EN-RAGE, and CASP-8 (Fig. 4). Other CCGF were more strongly expressed in plasma: CD8A, CD244, AXIN1, SCF, MCP-4, FGF-21, CCL19, IL12B, CCL23, and FGF19 (Fig.4). A group of CCGF were weakly expressed in both saliva and plasma with NPX ≤ ± 1.0 (MCP-3, IL-17 A, IL-2RB, IL-2, TSLP, IL-10RA, FGF-5, IL-15RA, ARTN, IL-20, IL-33, and IL-5) (Fig.4).

Table 3.

Comparison of eight biomarkers in plasma related surgical trauma and their expression in saliva

Saliva cytokine Plasma cytokine Uniprot Spearman(rs) p-value s/ns
IL6 IL6 P05231 0.153554 0.048937 s
CST5 CST5 P28325 −0.254204 0.000985 s
OSM OSM P13725 0.076021 0.331802 ns
TGF-alpha TGF-alpha P01135 0.104004 0.183712 ns
FGF-21 FGF-21 Q9NSA1 0.079302 0.311299 ns
IL10 IL10 P22301 0.078783 0.315976 ns
Flt3L Flt3L P49771 0.105294 0.178308 ns
FGF-19 FGF-19 O95750 0.023853 0.761040 ns
Open in a new tab

In an earlier study (Eriksson et al., submitted 2025) we found eight cytokines in plasma that significantly increased or decreased pre- versus postoperatively due to third molar surgical trauma; these were tested for correlations, against the same cytokines in saliva (s/ns = significant/not significant)

Fig. 3.

Fig. 3

Open in a new tab

Heat map. Pearson correlation among 8 inflammatory biomarkers in saliva (s) vs. plasma (p) (Eriksson et al., submitted 2025).

Fig. 4.

Fig. 4

Open in a new tab

Relative expression of 92 inflammatory biomarkers. Proteins in saliva vs plasma samples collected simultaneously from study participants. The differences between saliva and plasma are noticeably large in some cases, considering that they are presented on a log2 scale

Table 5.

Ninety-two inflammatory proteins (CCGF) expressed as median NPX (log2) value in saliva vs plasma. Every step by one unit on the NPX log2 scale means a doubling of the difference. Proteins showing the least difference in concentration (<±5%) between saliva and plasma are marked with (*)

Protein Saliva Median NPX
Log2 scale		 Plasma Median NPX
Log2 scale		 Least difference in concentration (< ± 5%)
saliva vs. plasma (*)
IL8 12.74 4.43
VEGFA 13.95 10.69
CD8A 1.92 9.52
MCP-3 −0.33 0.36
GDNF −0.05 1.32
CDCP1 7.58 2.07
CD244 0.20 4.76
IL7 1.75 0.83
OPG 8.56 8.85
LAP TGF-beta-1 4.38 5.49
uPA 9.25 9.28 *
IL6 5.48 1.84
IL-17 C 0.86 1.44
MCP-1 11.24 10.01
IL-17 A 0.66 0.03
CXCL11 4.49 6.81
AXIN1 1.16 5.50
TRAIL 9.50 7.98
IL-20RA 2.69 −0.19
CXCL9 7.30 5.54
CST5 8.72 6.30
IL-2RB −0.46 0.21
IL-1 alpha 10.16 −0.17
OSM 9.09 3.40
IL2 0.18 −0.68
CXCL1 11.29 8.15
TSLP −0.77 −0.36
CCL4 3.23 5.03
CD6 0.98 4.50
SCF 2.51 8.42
IL18 10.51 8.60
SLAMF1 −0.41 1.30
TGF-alpha 4.86 2.14
MCP-4 5.83 13.21
CCL11 1.13 6.25
TNFSF14 6.96 3.49
FGF-23 −0.65 1.02
IL-10RA −0.27 0.26
FGF-5 0.28 0.89
MMP-1 9.99 11.65
LIF-R 3.23 2.32
FGF-21 0.75 3.63
CCL19 4.93 9.53
IL-15RA 0.45 0.60
IL-10RB 3.25 5.19
IL-22 RA1 1.78 0.14
IL-18R1 7.28 5.91
PD-L1 3.12 4.80
Beta-NGF 0.70 1.22
CXCL5 12.13 10.52
TRANCE 2.04 4.53
HGF 8.33 7.10
IL-12B 2.10 5.53
IL-24 0.57 0.76
IL13 −1.14 −0.98
ARTN 0.05 −0.86
MMP-10 7.61 7.23
IL10 0.38 2.11
TNF 1.53 2.60
CCL23 3.19 10.83
CD5 4.89 4.80
CCL3 6.07 4.93
Flt3L 3.75 8.34
CXCL6 7.81 7.37
CXCL10 8.05 7.04
4E-BP1 4.44 7.48
IL-20 0.19 0.05
SIRT2 2.91 4.76
CCL28 6.58 1.06
DNER 9.91 7.81
EN-RAGE 7.48 1.58
CD40 10.06 10.11 *
IL33 0.80 0.28
IFN-gamma 5.83 6.24
FGF-19 1.41 7.42
IL4 −1.03 −0.34
LIF 1.87 −0.49
NRTN −0.34 −0.31 *
MCP-2 2.93 7.61
CASP-8 7.22 2.01
CCL25 0.41 5.17
CX3CL1 5.08 3.61
TNFRSF9 3.51 5.09
NT-3 0.96 2.67
TWEAK 6.02 8.60
CCL20 8.16 6.88
ST1A1 5.20 2.69
STAMBP 4.84 5.25
IL5 −0.33 −0.24
ADA 5.90 4.90
TNFB 0.78 3.89
CSF-1 8.78 8.83 *
Open in a new tab

Discussion

This extended study supports previous results, suggesting that the correlation between saliva and plasma regarding expression of inflammatory biomarkers is rather weak. The reasons why the concentrations of the same protein might differ between saliva and plasma are multiple. For example, saliva and plasma are separate compartments with different purposes. The expression of proteins in saliva may suffer from dilution effects due to stimulated and non-stimulated secretion and also influenced by the microenvironment in the mouth with its presence of microbes is different from plasma, which is supposed to be sterile. The method of analysing large numbers of protein from small sample volumes are highly specific and expressed in a relative strength of expression (NPX) which is accurate but lack the intuitive understanding of an absolute concentration. With that said comparisons between different matrix such as saliva and plasma are as correct as comparisons within one matrix [19, 20, 25]. This study found only four proteins with small differences (< ± 5%) in NPX-value between saliva and plasma; urokinase-type plasminogen activator (uPA), cluster of differentiation 40 (CD40), neurturin (NRTN), and colony-stimulating factor 1(CSF-1).

uPA is related to proteolysis in cancer invasion and metastasis [26]. CD40 is involved in regulation of B-cell responses and can also be expressed on dendritic cells, macrophages, fibroblasts and endothelial cells [27]. CD40 is involved in cell mediated immunity [28]. The expression of CD40 in the oral epithelium is restricted to keratinocytes and are inter- and intra-individually variated [28]. Neurturin is found to activate and sensitize bone afferent neurons inducing pain associated with bone pathology [29]. There are indications that CD40 as well as CSF-1 is elevated in painful diabetic neuropathy [30].

Prominent findings in saliva

This study showed that IL-6 in saliva and plasma and CST5 in saliva and plasma are both statistically significantly correlated, even though the correlations are weak. The study also showed 36 additional heterogeneous cytokine pair combinations to be significantly correlated between saliva and plasma, although again, the correlations remained weak. Nevertheless, some proteins were found to be much more strongly expressed in saliva, which may be the result of unknown patterns in the immune system. The most prominent proteins in saliva that we found were involved in a range of activities, such as angiogenesis, immune response and inflammation signalling, as well as apoptosis. IL-8 is a pro-inflammatory chemokine that primarily attracts and activates neutrophils [31]. The vascular endothelial growth factor A (VEGFA) promotes angiogenesis [32]. VEGFA regulates the formation of new blood vessels from existing ones [33]. IL-6 levels increase early (within an hour or faster) as a response to tissue damage caused by trauma or surgery, for instance [34]. The level of IL-6 is associated with the severity of the tissue damage [35]. Further, IL-6 has been shown to be essential in wound healing by modulating immune system activity [36]. IL-1 alpha is part of a group of proteins called alarmins. The group includes high-mobility group box 1 protein (HMGB1), IL-1 alpha, IL-33, and Ca2+ binding S100 [37]. The IL-1 family contains 11 members. In the development of inflammatory diseases and cancer, IL1-alpha plays a key role among the other alarmins [37]. IL-1 alpha is a dual function cytokine with both intracellular and extracellular functions. It is expressed in epithelial cells in the gastrointestinal tract, skin, liver, and kidney [37]. Release of IL-1 alpha leads to chemokine secretion with neutrophil infiltration [37]. Oncostatin M (OSM) is a part of the IL-6 family in which all members are important for cell signaling and cell communication in the immune system [38]. OSM is known to have protective effects on myelination, which is important for signal transduction along axons and for axonal protection [38]. TNF superfamily 14 (TNFSF14) is also known as LIGHT [39]. TNFSF14 is a transmembrane glycoprotein that plays a central role in the acute innate immune response [39]. TNFSF14 is involved in atherosclerosis and vascular inflammation [39]. CCL28 is a member of the chemokine family and activates immune cells throughout the body [40]. Due to its expression in epithelium and mucosal secretion, such as through milk and saliva, CCL28 may provide innate immune defence against a variety of bacterial pathogenes [40]. The highest levels of CCL28 expression can be found in salivary glands [40] and are increased in response to inflammatory events, such as rheumatoid [40]. RAGE is an immunoglobulin expressed in multiple tissues such as endothelium, vascular smooth muscle cells, and monocyte derived macrophages [41]. The activation of inflammatory cascades is a result of the binding of RAGE by EN-RAGE [41]. EN-RAGE facilitates inflammatory monocyte activation. EN-RAGE is a member of the S100 protein family [41]. Previous studies have observed increased levels of EN-RAGE in chronic inflammatory disorders. CASP-8 plays a role in initiating extrinsic apoptosis [42]. Apoptosis, programmed cell death [43], can be initiated by treatment with TNF, which activates CASP-8 [42].

Prominent findings in plasma

Some inflammatory biomarkers were more prominently expressed in plasma than saliva: CD8A, CD244, AXIN1, SCF, MCP-4, FGF-21, CCL19, IL-12B, CCL23, and FGF19 CD8/CD8 T-cell response are associated with the immune system´s antiviral response [44]. It is known that CD244 signalling correlates with certain virus persistence in humans, namely hepatitis B, hepatitis C, and tuberculosis (TB) [45]. In active human TB, CD244 signalling regulates repression of IFN-gamma and IFN alpha [45]. AXIN1 has been thought to be a tumour suppressor protein, although its function remains largely undefined [46]. SKP1-CUL1-F-box protein (SCF) has mostly been associated with cell proliferation, survival, and the connection to cancer. Disorders related to sleep, mood, metabolism, and intellect, need to be studied to better understand the function of SCF [47]. Monocyte chemoattractant protein (MCP)−4 is a pro-inflammatory protein overexpressed in many malignant tumours and may be important in the progression of tumours and metastasis [48]. Together with FGF-19 and FGF-23, fibroblast growth factor (FGF)−21 forms a subfamily [49]. FGF-21 is related to typ 2 diabetes, metabolic syndrome, coronary heart disease, obesity, and chronic kidney disease [50]. FGF-19 is reduced in diabetes and obesity and is inversely correlated to BMI shown in patients going through bariatric surgery [51, 52]. Chemokine receptor ligand (CCL)−19 promotes both breast- and cervical cancer progression. There are also indications that CCL19 suppresses lung cancer, colorectal cancer, ovarian cancer, and gastric cancer [53]. As a pro inflammatory cytokine, IL-12 is crucial for the antiviral immune response [54]. Both amount and variability of cytokine synthesis are genetically regulated by single nucleotide polymorphisms (SNPs) in IL-12 A and IL-12B; these also influence susceptibility to infectious diseases, disease severity, as well as the response to antiviral treatment [54]. Chemokine receptor ligand (CCL)−23 is associated with the outcome of stroke and acquired brain damage. Chemokines in the CC subgroup can attract basophils, monocytes, eosinophils, T lymphocytes, dendritic cells, and natural killer (NK) cells. CCL23 inhibits both production and release of monocytes and polymorphonuclear cells (PMNs) in bone marrow [55].

Biomarkers in saliva

Williams et al. compared two different samplings techniques for saliva and the levels of 27 cytokines in saliva and plasma, in 50 participants [1]. Williams et al. found that the correlations between saliva and plasma cytokine levels were not robust enough to allow substitution. As a result, they recommended that caution should be used in substituting saliva for plasma, and to consider that relationships can vary depending on the specific biomarker [1]. A study from 2021, of 43 adolescent competitive swimmers, which measured cytokines in saliva and plasma before and after exercise, states similar conclusions [2]. In this study, IL-6 levels in saliva and plasma differed significantly, with higher concentrations observed in saliva [2]. In another study, forty-eight different CCGF were analysed in saliva, plasma, and urine, in twenty healthy volunteers [4]. Thirty-seven of 48 CCGFs were found in plasma, 41 in saliva, and 34 in urine; this study reported the absolute concentration of each CCGF, but no correlation analysis was reported [4]. In a recent study of inflammatory biomarkers in children with juvenile idiopathic arthritis (JIA), saliva and serum were compared using the inflammatory panel from OLINK [56]. Cetrelli et al. concluded that the difference in biomarker patterns between saliva and serum is too large to justify using saliva as a substitute for plasma when assessing the degree of inflammation [56]. By analysing serum instead of plasma Cetrelli et al. offer a different perspective from other articles on the subject [56]. Serum lacks the coagulation factors still present in plasma, which makes the results not entirely comparable [18]. In a small study on children from 2021, IL-6, IL-10 and TNF-alpha were compared in saliva and plasma, which led to the conclusion that salivary measurement of cytokines is not a sufficient substitute for plasma, due to a weak correlation between these two fluids [2]. When investigating the inflammatory profile in patients with neuropathic pain, the most significant biomarkers for separating the groups was found in saliva as well in plasma and cerebrospinal fluid [57]. Jonsson et al. conclude that YKL-40 and MIP-1-alpha in saliva indicate a potential for further investigation [57]. Various authors have studied the possibility of using saliva as screening material for specific diseases, such as oral cancer [9], periodontitis [58], Alzheimer´s disease [59], amyotrophic lateral sclerosis (ALS) [60], cardiovascular disease [61], recurrent respiratory tract infections (rRTIs) in young children [62], and pregnancy outcome [63]. Majster et al. show increased IL-6 and MMP-10 levels in saliva in active inflammatory bowel disease [64]. It is well recognised that saliva and plasma differ in the expression of certain biomarkers. The strength of this study lies in its larger cohort and broader range of biomarkers, thereby confirming the conclusions of other authors [2, 18, 56].

Strength and limitations

The number of patients enrolled in this study is a strength, as is the fact that saliva and plasma were collected from the same individuals on the same occasion. The consequent procedure when collecting and preparing the biological samples is thus a strength. Another strength is the broad number of proteins analysed in comparison with similar, smaller studies [3]. All samples of saliva and plasma were handled consequently and according to a written routine to minimize the risk of any preanalytical error. There are some limitations with our study, however. First, this is a single site study. Second, there is an imbalance in sex representation, since 2/3 of the participants are women. Third, the study does not include children or elderly people, thus the age span is limited. Fourth, the analysis method (OLINK) is very sensitive and precise; for this reason, the results are reported as relative concentrations (NPX) rather than absolute concentrations (pg/ml). Fifth, no protease inhibitor was added to the saliva during sample collection, which means that some degree of proteolysis may have occurred in the samples. The exact consequence of not using protease inhibitors cannot be quantifies by our method. Sixth, although several biomarkers were studied, there could be numerous proteins of significant relevance to the topic that are not included in the selected panel. Seventh, the saliva and plasma samples were collected between 8 am and 2 pm, the intra individual range was only a few minutes. The inter individual differences may be caused by circadian variation, but we believe that the relatively large number of participants and the randomization procedures eliminate most of the risk of unrepresentative results. Lastly, since the study is cross-sectional, no conclusions regarding causality can be drawn and no conclusions can be drawn regarding physiological compliance between saliva and plasma.

Positive and negative control for unspecific binding awareness in the proximity extended assay (PEA) technology

Olink assays are developed with an emphasis on specificity and reducing non-specific binding. The cornerstone of the assay specificity is the dual antibody recognition and high-fidelity DNA-coupled measurement in the PEA protocol.

For each protein target, two oligonucleotide-coupled antibodies (PEA probes) must bind in close enough proximity to enable the oligos to hybridize and form a unique DNA template for detection. This overcomes the problems normally associated with multiplexed immunoassays,

The internal controls are spiked into every sample and are designed to monitor the three steps of the Olink protocol (immuno, extension and detection controls): Incubation control (immuno control) is non-human antigen that measured with PEA. Immuno control monitors potential technical variation in all three steps of the PEA reaction. The Extension Control is composed of an antibody coupled to a unique pair of DNA-tags. The Detection Control is a complete double stranded DNA amplicon which does not require any proximity binding or extension step to generate a signal. This control monitors the amplification/detection step.

Additionally, each sample plate contains a designated row of external controls. Sample control: Negative Control is also included in triplicate on each plate and consists of buffer run as a normal sample. These are used to monitor any background noise generated when DNA-tags come in close proximity without prior binding to the appropriate protein. The negative controls set the background levels for each protein assay and are used to calculate the limit of detection (LOD). Inter-plate Control (IPC) is included in triplicate on each plate, and these are run as normal samples. The median of the IPC triplicates is used to normalize each assay, to compensate for potential variation between runs and plates [19, 20, 25].

Conclusions

We found major differences in the levels of inflammatory biomarkers in saliva vs. plasma measured by the OLINK method. The main differences were that out of the ten most prominent proteins in saliva, six are especially interesting for further investigation. IL-6 and IL-8 are both pro-inflammatory cytokines. IL-6 and OSM belong to the same family; IL-1 alpha is part of another cytokine family and is anti-inflammatory. IL-6 is induced by IL-1 alpha, among several others. CST5 is a proteases inhibitor first found in saliva, and which modulates the whole inflammatory cascade by influencing the cytokine production. CCL28 is a chemokine with the highest level of expression in salivary glands. We also found inflammatory biomarkers in saliva that are weakly correlated with biomarkers in plasma: IL-6 and CST5. To summarize, saliva and plasma are not interchangeable when monitoring inflammatory biomarkers. This information may be important when planning for future inflammation studies.

Supplementary Information

Supplementary Material 1. (19.9KB, docx)

Acknowledgements

Not applicable.

Abbreviations

Alb

Albumin

AXIN

Axin

BMI

Body mass index

CASP

Caspase

CCGF

Cytokines, chemokines, and growth factors

CCL

Chemokine receptor ligand

CD

T-cell surface glycoprotein

CDCP

CUB domain-containing protein

OSM

Oncostatin M

CLCF

Cardiotrophin-like cytokine factor

CNTF

Ciliary neurotrophic factor

Crea

Creatinine

CSF

Macrophage colony-stimulating factor

CST5

Salivary cystatin D

CT

Creatine transporter

EDTA

Ethylenediaminetetraacetic acid

EN-RAGE

Protein S100-A12

EudraCT

European Union Drug Regulating Authorities Clinical Trials Database

EVF

Erytrocyte volume fraction

FasL

Fas Ligand

FGF

Fibroblast growth factor

Flt3L

FMS-related tyrosine kinase 3 ligand 

GCP

Good clinical practice

GDPR

General data protection regulation

IFN

Interferon

IL

Interleukin

IV

Intravenous

kD

kilo Dalton

LIF

Leukemia inhibitory factor 

LOD

Limit of detection

MCP

Monocyte chemoattractant protein

N

Number

NPX

Normalized protein expression

NRTN

Neurturin

PCR

Polymerase chain reaction

PDGF

Platelet derived growth factor

PEA

Proximity extension assay

Plt

Platelet count

RAGE

Receptor of advanced glycation end products

rs

Spearman rank coefficient

s/ns

Significant/not significant

SCF

Stem cell factor

SD

Standard deviation

TFG

Transforming growth factor

TGF

Tumour growth factor

TNF

Tumour necrosis factor

TNFSF

Tumour necrosis factor ligand superfamily member

uPA

Urokinase-type plasminogen activator

VEGFA

Vascular endothelial growth factor A

WBC

White blood cell count

Authors’ contributions

**LBE: ** Conceptualization, Methodology, Investigation, Data curation, Formal analysis, Visualization, Writing - original draft, Writing – review & editing, Project administration, Resources. **AL: ** Conceptualization, Methodology, Data curation, Formal analysis, Writing – review & editing. **ME: ** Writing – review & editing. **ÅT: ** Conceptualization, Methodology, Writing – review & editing, Resources. **AT: ** Writing – review & editing. **TG: ** Conceptualization, Methodology, Formal analysis, Writing – review & editing, Resources.

Funding

Open access funding provided by Uppsala University. The following financial support for the research, authorship, and/or publication of this article is acknowledged here: Uppsala-Örebro Regional Research Council, Public Dental Care Dalarna, Sweden, and the Centre for Clinical Research, Dalarna, Sweden.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Prior to any study procedure, we obtained informed consent from all patients. Good Clinical Practice (GCP), General Data Protection Regulation (GDPR), and the Helsinki Declaration were followed throughout the study.

The Swedish Ethical Review Authority revised and approved the study protocol (Reference number: 2015/378. Date: 02/12/2015).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  • 1.Williamson S, Munro C, Pickler R, Grap MJ, Elswick RK. Jr. Comparison of biomarkers in blood and saliva in healthy adults. Nurs Res Pract. 2012;2012:246178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Beigpoor A, McKinlay BJ, Kurgan N, Plyley MJ, O’Leary D, Falk B, et al. Cytokine concentrations in saliva vs. plasma at rest and in response to intense exercise in adolescent athletes. Ann Hum Biol. 2021;48(5):389–92. [DOI] [PubMed] [Google Scholar]
  • 3.Parkin GM, Kim S, Mikhail A, Malhas R, McMillan L, Hollearn M, et al. Associations between saliva and plasma cytokines in cognitively normal, older adults. Aging Clin Exp Res. 2023;35(1):117–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Khan A. Detection and quantitation of forty eight cytokines, chemokines, growth factors and nine acute phase proteins in healthy human plasma, saliva and urine. J Proteom. 2012;75(15):4802–19. [DOI] [PubMed] [Google Scholar]
  • 5.Loo JA, Yan W, Ramachandran P, Wong DT. Comparative human salivary and plasma proteomes. J Dent Res. 2010;89(10):1016–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Watkins LR, Maier SF, Goehler LE. Immune activation: the role of pro-inflammatory cytokines in inflammation, illness responses and pathological pain states. Pain. 1995;63(3):289–302. [DOI] [PubMed] [Google Scholar]
  • 7.Wautier JL, Wautier MP. Pro- and Anti-Inflammatory prostaglandins and cytokines in humans: A mini review. Int J Mol Sci. 2023;24(11). [DOI] [PMC free article] [PubMed]
  • 8.Chiamulera MMA, Zancan CB, Remor AP, Cordeiro MF, Gleber-Netto FO, Baptistella AR. Salivary cytokines as biomarkers of oral cancer: a systematic review and meta-analysis. BMC Cancer. 2021;21(1):205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Ishikawa S, Sugimoto M, Kitabatake K, Sugano A, Nakamura M, Kaneko M, et al. Identification of salivary metabolomic biomarkers for oral cancer screening. Sci Rep. 2016;6:31520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Bjornsson MA, Simonsson US. Modelling of pain intensity and informative dropout in a dental pain model after naproxcinod, Naproxen and placebo administration. Br J Clin Pharmacol. 2011;71(6):899–906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Quiding H, Jonzon B, Svensson O, Webster L, Reimfelt A, Karin A, et al. TRPV1 antagonistic analgesic effect: a randomized study of AZD1386 in pain after third molar extraction. Pain. 2013;154(6):808–12. [DOI] [PubMed] [Google Scholar]
  • 12.Rohatagi S, Kastrissios H, Sasahara K, Truitt K, Moberly JB, Wada R, et al. Pain relief model for a COX-2 inhibitor in patients with postoperative dental pain. Br J Clin Pharmacol. 2008;66(1):60–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Segerdahl M. Pain outcome variables–a never ending story? Pain. 2011;152(5):961–2. [DOI] [PubMed] [Google Scholar]
  • 14.Singla NK, Desjardins PJ, Chang PD. A comparison of the clinical and experimental characteristics of four acute surgical pain models: dental extraction, bunionectomy, joint replacement, and soft tissue surgery. Pain. 2014;155(3):441–56. [DOI] [PubMed] [Google Scholar]
  • 15.Eriksson LB, Gordh T, Karlsten R, LoMartire R, Thor A, Tegelberg A. Intravenous S-ketamine’s analgesic efficacy in third molar surgery. A randomized placebo-controlled double-blind clinical trial. Br J Pain. 2024;18(2):197–208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Eriksson LB, Gordh T, Karlsten R, Thor A, Tegelberg A. Patient safety of adjunct pre-operative intravenous S-ketamine for pain relief in third molar surgery - a randomised, placebo-controlled, double-blind trial. Br J Pain. 2024;18(6):450–60. 10.1177/20494637241262509. [DOI] [PMC free article] [PubMed]
  • 17.Huang J, Khademi M, Lindhe O, Jonsson G, Piehl F, Olsson T, et al. Assessing the preanalytical variability of plasma and cerebrospinal fluid processing and its effects on inflammation-related protein biomarkers. Mol Cell Proteomics. 2021;20:100157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.OLINK. Pre-analytical variation in protein biomarker research - White paper 2025 [updated 2020. Available from: https://7074596.fs1.hubspotusercontent-na1.net/hubfs/7074596/05-white%20paper%20for%20website/1095-olink-pre-analytical-variation-in-protein-biomarker-research-white-paper.pdf.
  • 19.OLINK. What is PEA? 2024 [Available from: https://olink.com/technology/what-is-pea.
  • 20.OLINK. User manual 2025 [Available from: https://olink.com/knowledge/documents?query=user%20manual.
  • 21.Assarsson E, Lundberg M, Holmquist G, Bjorkesten J, Thorsen SB, Ekman D, et al. Homogenous 96-plex PEA immunoassay exhibiting high sensitivity, specificity, and excellent scalability. PLoS One. 2014;9(4):e95192. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Spearman C. The proof and measurement of association between two things. By C. Spearman, 1904. Am J Psychol. 1987;100(3–4):441–71. [PubMed] [Google Scholar]
  • 23.Akoglu H. User’s guide to correlation coefficients. Turk J Emerg Med. 2018;18(3):91–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Chan YH. Biostatistics 104: correlational analysis. Singap Med J. 2003;44(12):614–9. [PubMed] [Google Scholar]
  • 25.OLINK. What is a good biomarker? 2025 [updated 2025-01-30. Available from: https://7074596.fs1.hubspotusercontent-na1.net/hubfs/7074596/6.Whitepapers/1594-Olink-White%20Paper%20What%20is%20a%20good%20biomarker%202025.pdf.
  • 26.Mahmood N, Mihalcioiu C, Rabbani SA. Multifaceted role of the Urokinase-Type Plasminogen Activator (uPA) and its receptor (uPAR): diagnostic, prognostic, and therapeutic applications. Front Oncol. 2018;8:24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Laman JD, De Boer M, Hart BA. CD40 in clinical inflammation: from multiple sclerosis to atherosclerosis. Dev Immunol. 1998;6(3–4):215–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Villarroel Dorrego M, Speight PM, Barrett AW. CD40 in human oral epithelia. Oral Oncol. 2007;43(7):626–33. [DOI] [PubMed] [Google Scholar]
  • 29.Nencini S, Ringuet M, Kim DH, Greenhill C, Ivanusic JJ. GDNF, neurturin, and artemin activate and sensitize bone afferent neurons and contribute to inflammatory bone pain. J Neurosci. 2018;38(21):4899–911. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Backryd E, Themistocleous A, Larsson A, Gordh T, Rice ASC, Tesfaye S, et al. Hepatocyte growth factor, colony-stimulating factor 1, CD40, and 11 other inflammation-related proteins are associated with pain in diabetic neuropathy: exploration and replication serum data from the Pain in Neuropathy Study. Pain. 2022;163(5):897–909. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Vilotic A, Nacka-Aleksic M, Pirkovic A, Bojic-Trbojevic Z, Dekanski D, Jovanovic Krivokuca M. IL-6 and IL-8: an overview of their roles in healthy and pathological pregnancies. Int J Mol Sci. 2022;23(23). [DOI] [PMC free article] [PubMed]
  • 32.Ke F, Xu MZ, Ma L, Chen QD, He BB, A JD. Progress and perspectives on BMP9-ID1 activation of HIF-1alpha and VEGFA to promote angiogenesis in hepatic alveolar echinococcosis. Front Oncol. 2024;14:1480683. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Kim M, Jang K, Miller P, Picon-Ruiz M, Yeasky TM, El-Ashry D, et al. VEGFA links self-renewal and metastasis by inducing Sox2 to repress miR-452, driving slug. Oncogene. 2017;36(36):5199–211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Desborough JP. The stress response to trauma and surgery. Br J Anaesth. 2000;85(1):109–17. [DOI] [PubMed] [Google Scholar]
  • 35.Ni Choileain N, Redmond HP. Cell response to surgery. Arch Surg. 2006;141(11):1132–40. [DOI] [PubMed] [Google Scholar]
  • 36.Hsing CH, Wang JJ. Clinical implication of perioperative inflammatory cytokine alteration. Acta Anaesthesiol Taiwan. 2015;53(1):23–8. [DOI] [PubMed] [Google Scholar]
  • 37.Bertheloot D, Latz E. HMGB1, IL-1alpha, IL-33 and S100 proteins: dual-function alarmins. Cell Mol Immunol. 2017;14(1):43–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Houben E, Hellings N, Broux B. Oncostatin M, an underestimated player in the central nervous system. Front Immunol. 2019;10:1165. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Ea HK, Kischkel B, Chirayath TW, Kluck V, Aparicio C, Loeung HU, et al. Systemic inflammatory cytokine profiles in patients with gout during flare, intercritical and treat-to-target phases: TNFSF14 as new biomarker. Ann Rheum Dis. 2024;83(7):945–56. [DOI] [PubMed] [Google Scholar]
  • 40.Mohan T, Deng L, Wang BZ. CCL28 chemokine: an anchoring point bridging innate and adaptive immunity. Int Immunopharmacol. 2017;51:165–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Ligthart S, Sedaghat S, Ikram MA, Hofman A, Franco OH, Dehghan A. EN-RAGE: a novel inflammatory marker for incident coronary heart disease. Arterioscler Thromb Vasc Biol. 2014;34(12):2695–9. [DOI] [PubMed] [Google Scholar]
  • 42.Tummers B, Green DR. Caspase-8: regulating life and death. Immunol Rev. 2017;277(1):76–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Mo W, Su S, Shang R, Yang L, Zhao X, Wu C, et al. Optogenetic induction of caspase-8 mediated apoptosis by employing Arabidopsis Cryptochrome 2. Sci Rep. 2023;13(1):23067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Loucif H, Dagenais-Lussier X, Beji C, Cassin L, Jrade H, Tellitchenko R, et al. Lipophagy confers a key metabolic advantage that ensures protective CD8A T-cell responses against HIV-1. Autophagy. 2021;17(11):3408–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Wang Y, Zhong H, Xie X, Chen CY, Huang D, Shen L, et al. Long noncoding RNA derived from CD244 signaling epigenetically controls CD8 + T-cell immune responses in tuberculosis infection. Proc Natl Acad Sci U S A. 2015;112(29):E3883-92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Mazzoni SM, Fearon ER. AXIN1 and AXIN2 variants in gastrointestinal cancers. Cancer Lett. 2014;355(1):1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Skaar JR, Pagan JK, Pagano M. SCF ubiquitin ligase-targeted therapies. Nat Rev Drug Discov. 2014;13(12):889–903. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Li S, Hu Y, Liu O, Li X, Lin B. Prognostic biomarker MCP-4 triggers epithelial-mesenchymal transition via the p38 MAPK pathway in ovarian cancer. Front Oncol. 2022;12:1034737. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Fisher FM, Maratos-Flier E. Understanding the physiology of FGF21. Annu Rev Physiol. 2016;78:223–41. [DOI] [PubMed] [Google Scholar]
  • 50.Shayota BJ. Biomarkers of mitochondrial disorders. Neurotherapeutics. 2024;21(1):e00325. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Ryan PM, Hayward NE, Sless RT, Garwood P, Rahmani J. Effect of bariatric surgery on circulating FGF-19: a systematic review and meta-analysis. Obes Rev. 2020;21(8):e13038. [DOI] [PubMed] [Google Scholar]
  • 52.Somm E, Jornayvaz FR. Fibroblast growth factor 15/19: from basic functions to therapeutic perspectives. Endocr Rev. 2018;39(6):960–89. [DOI] [PubMed] [Google Scholar]
  • 53.Zhou R, Sun J, He C, Huang C, Yu H. CCL19 suppresses gastric cancer cell proliferation, migration, and invasion through the CCL19/CCR7/AIM2 pathway. Hum Cell. 2020;33(4):1120–32. [DOI] [PubMed] [Google Scholar]
  • 54.Benmansour R, Tagajdid MR, Lahlou IA, Oumzil H, El Hamzaoui H, Fjouji S, et al. Implication of IL-12A, IL-12B, IL-6, and TNF single-nucleotide polymorphisms in severity and susceptibility to COVID-19. Int J Immunopathol Pharmacol. 2024;38:3946320241279893. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Faura J, Bustamante A, Penalba A, Giralt D, Simats A, Martinez-Saez E, et al. CCL23: a chemokine associated with progression from mild cognitive impairment to Alzheimer’s disease. J Alzheimers Dis. 2020;73(4):1585–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Cetrelli L, Lundestad A, Gil EG, Fischer J, Halbig J, Frid P, et al. Serum and salivary inflammatory biomarkers in juvenile idiopathic arthritis-an explorative cross-sectional study. Pediatr Rheumatol Online J. 2024;22(1):36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Jonsson M, Gerdle B, Ghafouri B, Backryd E. The inflammatory profile of cerebrospinal fluid, plasma, and saliva from patients with severe neuropathic pain and healthy controls-a pilot study. BMC Neurosci. 2021;22(1):6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Kinney JS, Morelli T, Braun T, Ramseier CA, Herr AE, Sugai JV, et al. Saliva/pathogen biomarker signatures and periodontal disease progression. J Dent Res. 2011;90(6):752–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Ashton NJ, Ide M, Zetterberg H, Blennow K. Salivary biomarkers for Alzheimer’s disease and related disorders. Neurol Ther. 2019;8(Suppl 2):83–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Sjoqvist S, Otake K. Saliva and saliva extracellular vesicles for biomarker candidate Identification-Assay development and pilot study in amyotrophic lateral sclerosis. Int J Mol Sci. 2023;24(6). [DOI] [PMC free article] [PubMed]
  • 61.Yousif G, Murugesan S, Djekidel MN, Terranegra A, Gentilcore G, Grivel JC, et al. Distinctive blood and salivary proteomics signatures in Qatari individuals at high risk for cardiovascular disease. Sci Rep. 2025;15(1):4056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Koenen MH, de Steenhuijsen Piters WAA, de Jonge MI, Langereis JD, Nierkens S, Chu M et al. Salivary polyreactive antibodies and haemophilus influenzae are associated with respiratory infection severity in young children with recurrent respiratory infections. Eur Respir J. 2024;64(4). [DOI] [PMC free article] [PubMed]
  • 63.Guterstam YC, Acharya G, Schott K, Bjorkstrom NK, Gidlof S, Ivarsson MA. Immune cell profiling of vaginal blood from patients with early pregnancy bleeding. Am J Reprod Immunol. 2023;90(2):e13738. [DOI] [PubMed] [Google Scholar]
  • 64.Majster M, Lira-Junior R, Hoog CM, Almer S, Bostrom EA. Salivary and serum inflammatory profiles reflect different aspects of inflammatory bowel disease activity. Inflamm Bowel Dis. 2020;26(10):1588–96. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 1. (19.9KB, docx)

Data Availability Statement

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Articles from BMC Oral Health are provided here courtesy of BMC

RESOURCES