Introduction : The transplantation field requires specific biomarkers to assess the level of immunosuppression. We aimed to analyze the independent correlation between the number of circulating lymphocytes, simply and routinely monitored by blood cell counts during outpatient visits, and patient and graft survival to explore its role as a potential biomarker of immunosuppression. Methods : 3002 kidney or combined kidney-pancreas transplanted patients between January 2000 and December 2016, from two University Hospitals, alive with a functioning graft at 1 year post-transplantation, were enrolled. Clinical and biological information were extracted from the DIVAT data base. We investigated the etiological relationship between time-dependent lymphocyte count after one year of transplantation and patient and graft survivals, viral infection and cancer risks using a time-dependent multivariate Cox model. Results : A patient with a lymphocyte count below 750 /mm3 at a given time within the follow-up had a higher risk of graft failure (HR 3.08, p<0.001) and death (HR 2.06, p<0.001) when compared to a similar patient with a normal lymphocyte count (more than 1500 /mm3) at the same time, independently from other classical confounding factors. Patients with less than 750 /mm3 lymphocytes were more at risk of viral infections than comparable patients with a normal lymphocyte count (HR 1.62, p<0.001). Conclusion : Deep lymphopenia over time is highly associated with a risk of graft failure, death and occurrence of viral infection. These data suggest that the longitudinal lymphocyte count could be used as a simple routine biomarker of long-term graft and patient outcome.