Genomic diversity of endogenous FeLV in the Felis genus: findings and limitations

Annals of the New York Academy of Sciences · 2026

In plain language

This study examined feline genomes to describe endogenous feline leukemia virus sequences, which are present in their DNA; it did not study a clinical group of cats. The authors found fewer regions of the viral gene pol than of gag or env, with a statistically significant trend toward loss of pol. The effects of specific endogenous viral insertions on the body, and how they interact with FeLV acquired through infection, remain unknown. These findings do not identify any insertion as a cause or predictor of feline lymphoma or leukemia.

Scientific detail

This computational study, published in Annals of the New York Academy of Sciences in 2026 (doi: 10.1111/nyas.70278), characterizes the genomic diversity of endogenous feline leukemia virus (enFeLV) in the Felis genus. The abstract reports 17 felid genomes. The methods describe a family-level genome dataset and a second dataset containing available chromosome-level Felis assemblies plus Prionailurus bengalensis, without specifying an overall numerical total. The results report 141 enFeLV copies in long-read genome assemblies and 65 typical copies used for insertion dating. These are genomic datasets and copy counts, not a clinical cohort.

The authors found fewer pol regions than gag or env regions, with a statistically significant trend of pol deletion. This finding supports unequal retention of viral gene regions in the annotated sequences but does not demonstrate an effect on cancer risk. The relevant passage describes all 65 copies as nonrecombinant, conflicting with an earlier result reporting recombination in seven typical copies and leaving the precise analysis subset unclear.

As biological background, the introduction identifies recombination between endogenous and exogenous FeLV, particularly in the envelope region, as the origin of FeLV-B, a subgroup described as having increased oncogenic potential. However, the study does not test FeLV-B generation, lymphoma or leukemia development, or the oncogenic effects of specific endogenous loci. Predicted open reading frame (ORF) integrity and endogenous recombination signals do not establish tumor causation. The effects of specific enFeLV loci on host biology and their interactions with exogenous FeLV remain unresolved.

The authors caution that short-read assemblies may introduce artifacts and fail to resolve polymorphic loci; one assembly combined multiple individuals. Genomes of four Felis species were unavailable, limiting host coverage, and haploid assemblies prevented assessment of heterozygosity. Insertion dating assumes a stable mutation rate and neutral evolution, neither established for enFeLV. This approach cannot precisely date insertions with identical long terminal repeats (LTRs), and recombination may bias the estimates. Functional studies are still needed to establish biological consequences.

What we know

In the annotated sequences, the authors found fewer pol regions than gag or env regions, with a statistically significant trend of pol deletion. This supports unequal retention of viral gene regions, not a demonstrated effect on cancer risk. An internal discrepancy concerning the nonrecombinant status of the 65 copies described leaves the precise analysis subset unclear.

What we don't know

The effects of specific enFeLV loci on host biology and their interactions with exogenous FeLV have not been established. This study does not determine whether a particular endogenous insertion causes or predicts lymphoma or leukemia, nor does it test FeLV-B generation. Functional studies are lacking, and limitations in assemblies, species coverage, and dating assumptions constrain interpretation of the findings.

Evidence

Provenance

This project does not diagnose or prescribe. Always consult your veterinarian.