Cross-protection is a biocontrol method to protect plants from damage caused by pathogenic viruses. This strategy of cross-protecting a plant with a mild strain to prevent subsequent infection by related severe strains was first described in 1926. Although cross-protection was first evidenced nearly a century ago, the underlying mechanism(s) remain(s) poorly understood. In particular, although the genetic relatedness between cross-protective and challenge strains has been emphasized in a number of papers, the percentage of identity required for cross-protection remains unknown. This lack of data has led to a long and sometimes unsuccessful empirical search for cross-protective strains against some viral pathogenic ones. Recent papers have shown that cross-protection occurs only among variants of the same strain and not between variants of different strains for the citrus tristeza virus (Closterovirus) and the pepino mosaic virus (Potexvirus). In order to determine the percentage of identity required between the cross-protective and the challenge strains for cross-protection to be effective against the grapevine fanleaf virus (GFLV, Secoviridae), different cross-protective variants with decreasing sequence homology with respect to the challenge variant were selected and inoculated on Nicotiana benthamiana. Our results suggest that in addition to the percentage of identity, the entry point of the cross-protective and challenge strains impacts the cross-protection success.