01Vector Choice
A gene therapy trial needs a vector that infects both dividing and non-dividing cells, carries up to 8 kb of transgene, and does NOT integrate into the host genome. Which virus?
Adenovirus. It is the most commonly used viral vector in gene therapy. Its DNA stays episomal (does NOT integrate), so no insertional mutagenesis risk. Infects dividing AND non-dividing cells. Drawback: triggers a strong innate immune response that limits repeat dosing.
02Vaccine Strategy
The J&J COVID vaccine, AstraZeneca, and Sputnik V all use adenovirus vectors. But Sputnik used TWO different serotypes (Ad26 then Ad5) for its two doses. Why not use the same serotype twice?
Because dose 1 generates anti-vector immunity against that serotype. If dose 2 uses the same serotype, the immune system destroys the vector before it can deliver its payload. Switching serotypes dodges this. AstraZeneca used chimpanzee adenovirus (ChAdOx1) so most humans had no pre-existing immunity at all. All three vaccines deliver the SARS-CoV-2 spike gene inside a replication-deficient adenovirus (E1/E3 genes deleted).
03Oncolytic
A clinical trial uses an engineered adenovirus that selectively replicates in p53-deficient tumor cells, lysing them while sparing normal tissue. What is this approach called?
Oncolytic virotherapy. The virus is designed to only replicate where p53 is mutated (i.e., cancer cells). ONYX-015 was among the first tested. China approved Oncorine (H101) in 2005 for head and neck cancer. T-VEC (HSV-based) proved the concept works in melanoma. The idea: weaponize a virus to eat the tumor from inside.
04Safety · 1999
In 1999, an 18-year-old with OTC deficiency received a high-dose IV injection of an Ad5 vector. Within 4 days he was dead. What killed him, and what did it change?
Massive cytokine storm from the adenoviral vector. Jesse Gelsinger's death at UPenn led to a complete FDA overhaul of gene therapy oversight. Key lesson: adenovirus vectors trigger strong innate immunity, especially at high IV doses. This is why newer gene therapy vectors use AAV (adeno-associated virus, a parvovirus) instead, which is far less immunogenic.
05Board Synthesis
Board synthesis: A question lists 5 viral vectors and asks which integrates into the host genome. Rank them: adenovirus, AAV, retrovirus, lentivirus, and adenovirus-associated satellite.
Integrates: Retrovirus (random integration via integrase = insertional mutagenesis risk, only divides cells). Lentivirus (HIV-derived, also integrates, but infects non-dividing cells too).
Does NOT integrate: Adenovirus (episomal, strong immune response). AAV (episomal, low immunogenicity, small capacity ~4.7 kb). Board trap: AAV can rarely integrate at AAVS1 site on chromosome 19, but this is not considered clinically significant.
Does NOT integrate: Adenovirus (episomal, strong immune response). AAV (episomal, low immunogenicity, small capacity ~4.7 kb). Board trap: AAV can rarely integrate at AAVS1 site on chromosome 19, but this is not considered clinically significant.
06Biostats Bonus
Biostats bonus: A phase III gene therapy trial using an AAV vector reports 3 cases of hepatotoxicity in 200 treated patients vs 0 in 200 placebo patients. The p-value is 0.12. The FDA advisory committee votes 11-6 to approve. What concepts are at play?
The hepatotoxicity signal is not statistically significant (p = 0.12 > 0.05), but the trial may be underpowered to detect a rare adverse event (NNH = 67). The split vote suggests the committee weighed clinical significance vs statistical significance. With rare diseases, trials are small, and a type II error (false negative) for safety signals is a real risk. Post-market surveillance (phase IV) and REMS programs can catch what the trial missed. This is exactly what happened with early gene therapy vectors.
