Primary Sjögren's syndrome (pSS) is a mild chronic systemic autoimmune disease characterised by a slow progression and low morbidity and mortality rates [¹,²]. Nonetheless, despite a generally rather indolent course, patients with pSS are widely known to have an increased risk for developing B-cell non-Hodgkin's lymphoma (B-cell NHL) [³]. To date, many efforts have been made in order to identify early predictors of B-cell NHL in pSS, but it is still difficult to identify who among the patients with pSS will have a progression to lymphoma [⁴]. In the last few years, proteomic analysis has been increasingly applied to the study of whole saliva in pSS in order to identify novel diagnostic biomarkers for the disease [^5-7]. Only few studies have been carried out on the modification of salivary proteomic profiles during and after pharmacological treatments [⁸]. In this study, we used a proteomic approach to identify salivary therapeutic biomarkers in a patient with pSS and B-cell NHL-MALT type of the salivary glands, successfully treated with cyclophosphamide (CTX) and rituximab (RTX). Proteomic analysis of the patient's whole saliva, performed before, during the treatment course, and after 6 months of pharmacological therapy, showed an intriguing correspondence between the patient's clinical improvement and the changes of her proteomic salivary profile, highlighting a potential application of proteomics in identifying therapeutic biomarkers and in monitoring the efficacy of pharmacological treatments.

Figures 2 and 3 clearly illustrates the differences in the patient's salivary profile at baseline (M1), after the fourth infusion of RTX (M2) and after 6 months with the patient on remission (M3). Overall, 18 spots appeared to be differently expressed in the salivary profile of the patient at baseline and during treatment in comparison with the proteomic pattern obtained on remission. Mass spectrometry identification showed that these spots collapsed essentially on six main proteins, including putative uncharacterized albumin (spots n° 141, 146, 147, 249, 251, 274, 294, 297, 325), Immunoglobulin J chain (spots n°243, 244, 259, 265), Ig kappa chain C region (spot n° 218), alpha-1-antitrypsin (spots n°132, 134), Ig alpha-1 chain C region (spot n° 140) and haptoglobin (spot n° 329). A list of identified proteins, MW, pI, score and coverage values of MS/MS is shown in Table 2. From a qualitative point of view, when we compared the M1 and M3 profiles we observed a complete disappearance of many spots related to Ig alpha-1 chain C region, Ig kappa chain C region, Immunoglobulin J chain and putative uncharacterized protein albumin. From a quantitative point of view, the most relevant differences between M1 and M3 were detected for haptoglobin (5.8-fold decrease, p-value = 0.0019) and alpha-1-antitrypsin (6.3-fold decrease, p-value = 0.036). These two proteins were also characterised by a significant change in the expression level after the fourth infusion of RTX with haptoglobin and alpha-1-antitrypsin showing a 2.6-fold (p-value = 0.011) and a 2.1-fold (p-value = 0.05) decrease respectively, when compared to the baseline.

Recent advances in proteomic technologies have permitted the discovery of several salivary candidate biomarkers for the diagnosis of pSS [^13-15]. So far, the potential usefulness of salivary proteomics for the discovery of therapeutic biomarkers has been scarcely investigated. Nonetheless, preliminary results have shown a modification of pSS proteomic profile before and after pilocarpine [⁸] This study, for the first time, demonstrated a strong correspondence between the salivary proteomic profile and the clinical disease course in a patient with pSS and B-cell NHL-MALT type of the salivary glands. In particular, at the baseline, we described the over-expression of seven different spots collapsing into immunoglobulin J chain, IgK chain C region and Ig alpha-1 chain C region, which could be directly correlated to the lymphoproliferative process [¹⁶]. In addition, we also demonstrated a significant increase of the optical density of the spots related to haptoglobin and alpha-1-antitrypsin, and the presence of nine spots subsequently identified as albumin fragments. It is noteworthy that increased levels of serum haptoglobin and alpha-1-antitrypsin have been already described in patients with several different haematological and non-haematological malignancies, including pancreatic cancers, hepatocellular carcinoma, oral cancers, thyroid carcinomas, and non-small cell lung cancer [^17-23]. In particular, an increase of the 21 kDa haptoglobin-related protein (Hpr) (which was the same protein identified in the whole saliva of our patient) has been reported in the serum of patients with malignant lymphoma, with advanced disease and "B" symptoms [²³]. Moreover, numerous fragments of albumin and of alpha-1-antitrypsin were detected in the urine of children with cancer [²⁴,²⁵]. To our knowledge, this is the first time that these proteins are described in the whole saliva of a patient with pSS and B-cell NHL. In our case, we also found that Hpr and alpha-1-antitrypsin significantly decreased in response to treatment, suggesting a potential role of these proteins as early therapeutic biomarkers in the clinical setting of lymphoma.

Overall, the patient's salivary profile and its modification over-time represent an indirect proof of the capacity of whole saliva to reflect systemic conditions. Moreover, in this case we demonstrated a specific correspondence between clinical improvement and proteomic changes of the salivary peptide complex, shedding new light on the potential usefulness of proteomic analysis in discovering not only diagnostic but also prognostic and therapeutic biomarkers in patients' whole saliva. We might therefore speculate that, during the follow-up of patients with lymphomas, proteomic analysis might be able to detect salivary biomarkers early predictors of treatment response.

Overall, once validated in larger studies, these biomarkers could improve patients' management into clinical daily care and facilitate patients' prognostic classification. From the perspective of the research, the analysis of biomarkers signatures in saliva could also help to clarify the pathogenetic pathways underlying lymphoproliferation in pSS leading to develop new concepts for early diagnosis and curative therapies.

The authors declare that they have no competing interests.

CB, LG and AL designed the study, coordinated the research, analyzed data and wrote the manuscript; SB and LB participated in the design and coordination of study and helped to draft the manuscript; CG carried out statistical analysis; FC, YD and ED carried out proteomic analysis; CB, FF and SG carried out pharmacological treatment; VD carried out the immunohystochemical analysis. All authors read and approved the final manuscript.