The clinical documentation comprises 9 phase 1 studies, 4 phase 2 dose-finding studies, 2 phase 3

studies and an extension study. Additionally 15 studies performed in China, some with a comparable

but not the same formulation and others containing varying levels of DDAIP were included. Some of

the studies do not contain the uptake enhancer DDAIP, consequently the doses to be administered

are higher (up to 1000 µg).

Fifteen studies were performed in China most with a comparable but not the same formulation. Some

of the alprostadil formulations studied did not contain the uptake enhancer DDAIP, consequently the

doses to be administered are higher (up to 1000 µg), while other studies evaluated various levels of

DDAIP. Given the lack of equivalence between the formulation applied and the Chinese formulation,

and because the Chinese formulation is administered at higher doses (up to 1000 µg) than the

formulation applied for (200 or 300 µg) and there are some differences in the formulation excipients,

results obtained in the Chinese studies cannot be extrapolated. Results are therefore not reported.

Two carcinogenicity studies have been completed on DDAIP including a 26-week dermal application

in Tg.AC mice and a 2-year subcutaneous dosing study in rats. Two other carcinogenicity studies

were completed on DDAIP HCl including a dermal study in mice and a dermal study of terbinafine HCl

Nail Lacquer (containing 0.5% DDAIP HCl) in rats. The transgenic mouse study, using a model

specifically sensitive to dermally applied carcinogens, and used in this way several times for regulatory

purposes, was unexpectedly positive, and DDAIP has been shown to induce papilloma’s after dermal

application. The other three studies were negative.

DDAIP has a similarity to cationic surfactant lauric acid diethanolamine (LADA), sharing with DDAIP

the lauryl (C12) tail, and therefore its detergent action. Also LADA was tested in the TG.AC mouse

and reported to be positive. Therefore the following points were discussed to come to a risk

assessment for DDAIP:

1. Extensive use of LADA for more than 25 years in consumer products including those that are

considered ‘leave-on’ products and expose mucous membranes support the safety and lack of

tumorigenicity of this compound at concentrations up to 9%.

2. A survey of US approved drugs illustrated that a number of both prescription drugs and over-thecounter

products tested positive in the Tg.AC transgenic mouse model.

Both LADA and DDAIP tested positive in the Tg.AC transgenic model. Papilloma formation in Tg.AC

mice is positively correlated with irritation at the site of application. LADA and DDAIP are both

detergents, and due to this characteristic this will probably lead to similar irritation. Overall, it can be

concluded that the papilloma-inducing effect of DDAIP is caused by the irritation in this TG.AC mouse

model, and is unlikely to be of human relevance.