Introduction:

Surveillance biopsy remains debated after renal transplantation since the histological diagnostic do not lead to clear therapeutic recommendations. The main objective for this study was to identify and validate a clinical diagnostic signature of histological lesions from surveillance biopsy at one year after renal transplantation.

Patients and methods:

From the DIVAT cohort, we studied 567 adult recipients of kidney or kidney-pancreas, transplanted between 2006 and 2012, from deceased donor, and alive with a functioning graft at one year post-transplantation. Patients displaying major histological lesions comprising “allo-immunity” (i.e. Borderline, acute or chronic humoral and/or cellular rejection), IFTA grade 2 or 3 and initial disease relapse were compared with patients presenting minor lesions (i.e. Normal histology or IFTA grade 1). The diagnostic signature was identified using a lasso penalized logistic regression and internally validated by ROC 0.632+ bootstrap.

Results:

303 patients (53%) were diagnosed with minor histological lesions. The diagnostic signature included 8 variables. Women recipients (OR=2.70), patients with a kidney transplant alone (OR=1.36), with high serum creatinine at 3 months (OR=1.05), 6 months (OR=1.58) and 12 months (OR=1.42), receiving a male kidney (OR=1.48), with anti-class II immunization at transplantation time (OR=1.47) and who experienced dialysis before transplantation (OR=2.81) are more at risk of major histological lesions. The area under the associated ROC curve is estimated at 0.69. We assumed a minimal negative predictive value at 75% leading to a discriminating threshold. In this medical decision making perspective, 17% of patients have a good chance to display minor lesions and a surveillance biopsy should not be proposed.

Conclusion:

This diagnostic signature can help physicians to not recommend surveillance biopsies in patients at high risk of minor lesions and therefore avoid unnecessary biopsies.

« Back to 2015 American Transplant Congress