Gates Package — page 163

at the enterprises now investigating vaccine candidates to ascertain their intended mode of manufacture. Manufacturing plans could be made accordingly, with a great deal initiated immediately. Drug companies frequently build production capability before they have an approved product. How is this different? The principle is the same and the precedent is instructive. But this effort would be unprecedented in scale, scope and timing. The scale required is a production capacity for two to three hundred million doses for just the United States and more than two to three billion doses globally. Moreover, a drug company knows the specific products and processes for which it is building. This effort will have to prepare for manufacture of at least the leading 10 or so candidates and perhaps for many more. Finally, to meet a six-month target, investments will have to be made much earlier in the development cycle than is customary for companies. Can production capacity be found or created at the scale required? Some can be created and for an RNA vaccine an entirely new infrastructure would be required. For other types of vaccine, existing infrastructure would have to be reallocated. Some unused capability is readily identifiable. A majority of capacity, however, is likely to come from reallocation of existing facilities from a significant number of companies. Reallocation requires planning and discussion about financial costs and risks (e.g., if flu vaccine capabilities were reallocated to COVID-19). CEPI (Richard Hatchett) has pointed out that early recognition of potential reallocation demands could have the benefit of stimulating some present product to 24/7 production, thereby softening the effects of later reallocation. Discussion and planning is also necessary to establish a regime for sharing IP, financial reward, etc. and for standardizing production. It seems evident, however, that the capacity is there. Is it possible to assess the safety and efficacy of a vaccine candidate within six months? It is possible, though not by any means certain. Meeting that target might involve a month for determining a compound and manufacturing it in small quantities for initial trials; two months for initial trials (perhaps a combined Phase 1 and Phase 2) titrating the compound by administering it in varying doses and analyzing blood serum results; and three months for a cohort study or challenge study in lieu of a typical Phase 3 trial. The resulting vaccine might only be approved for certain populations pending a bridge study of, for example, other age groups and a Phase 4 study drawn from evaluation of those who are inoculated. How would a cohort study work? Inoculation would be provided to a subset of health care workers in some high attack environments, while a control group would receive a placebo. Statistical analysis would compare results from exposure. How would human challenge trials work? Are they morally defensible? We have substantial experience with challenge trials. Volunteers would have to be informed of the risks and probably would be concentrated in age ranges where risks were lowest.