Predicting the risk for lymphoma development in Sjogren syndrome: An easy tool for clinical use

Abstract

The heightened risk of non-Hodgkin lymphoma (NHL) development in primary Sjogren syndrome (SS) is well established. Several adverse clinical and laboratory predictors have been described. In the current work, we aimed to formulate a predictive score for NHL development, based on clinical, serological, and histopathological findings at the time of SS diagnosis. In the present case-control study of 381 primary SS patients and 92 primary SS patients with concomitant NHL, clinical, serological, and histopathological variables at the time of SS diagnosis were retrospectively recorded. For the identification of predictors for NHL development univariate and multivariate models were constructed. Salivary gland enlargement (SGE), lymphadenopathy, Raynaud phenomenon, anti-Ro/SSA or/and anti-La/SSB autoantibodies, rheumatoid factor (RF) positivity, monoclonal gammopathy, and C4 hypocomplementemia were shown to be independent predictors for NHL development. On the basis of the number of independent risk factors identified, a predictive risk score for NHL development was formulated. Thus, patients presenting with ≤2 risk factors had a 3.8% probability of NHL development, those with 3 to 6 risk factors 39.9% (OR (95%CI): 16.6 [6.5-42.5], P < 0.05), while in the presence of all 7 risk factors the corresponding probability reached 100% (OR [95%CI]: 210.0 [10.0-4412.9], P < 0.0001). In conclusion, an easy to use diagnostic scoring tool for NHL development in the context of SS is presented. This model is highly significant for the design of early therapeutic interventions in high risk SS patients for NHL development.

Figure 1

Identification of independent risk factors for NHL development in 2 steps. First, risk factors identified to be statistically significant in the univariate analysis were analyzed in 3 separate multivariate models and independent clinical, laboratory, and histopathological risk factors for NHL development were determined. Second, a final multivariate logistic regression analysis including all independent risk factors, revealed in the 3 separate multivariate models, was performed. MSG = minor salivary gland, NHL = non-Hodgkin lymphoma, PNS = peripheral nervous system, RF = rheumatoid factor, SGE = salivary gland enlargement.

Figure 2

The performance evaluation of the predictive model for NHL development with the formation of ROC curves. The AUC was 0.9 (95%CI: 0.8–0.9, P < 0.001). AUC = area under the curve, CI = confidence interval, NHL = non-Hodgkin lymphoma, ROC = receiver operating characteristic.

Figure 3

The probability of NHL development among SS patients was estimated on the basis of the number of independent risk factors. The probability of NHL development was 3.8% for patients presenting with ≤2 risk factors, 39.9% for those displaying 3 to 6 risk factors, and 100% in the presence of all 7 risk factors. The OR along with the corresponding CI and P-values for NHL development in the presence of all 7 risk factors were 210.0 (10.0–4412.9), P < 0.0001 compared to those with 2 or less risk factors. The corresponding values in the presence of 3 to 6 risk factors were 16.6 (6.5–42.5), P < 0.05 in comparison with those with 2 or less risk factors. CI = confidence interval, NHL = non-Hodgkin lymphoma, OR = odds ratio, SS = Sjogren syndrome.