Epidemiology and HIV’s Physical Properties

OCT.19 '95 7:34AM 310-446-1664 LAYNE @ UCLA. [phone] P.6
COMMENTARY
potent antiviral agents have been initiated, which promise to reduce the probability of neonatal
transmissions even further.
The factors influencing the spread of HIV from mother to infant may provide crucial insights
into other modes of transmission. In particular, associations between transmissibility and viral
load suggest that more infectious individuals carry greater loads in their circulation. The analyses
of antiviral drug trials by mathematical models demonstrate that in vivo concentrations of HJV
obey so-called "steady-state" kinetics, where clearance of infection by the immune system matches
ongoing viral replication1118. This kinetic behavior means that HIV isolates with larger
"reproductive numbers" - the total number of infectious virions generated by each infected cell -
generate correspondingly larger viral loads and vice versa (Fig. 2a). Infected individuals also
harbour swarms of HIV isolates in their blood called "quasispecies" which possess closely related
genotypes yet exhibit broadly divergent phenotypes. Each unique member of this quasispecies
may have different reproductive numbers, cellular tropisms and immune clearance rates11. With
these dynamic relationships, it is conceivable that viral loads and reproductive numbers may act
alone or in concert to affect person-to-person transmissibility (Fig. 2b).
Measurements of the infectious fraction, gpl20 shedding, reverse transcriptase loss and
reproductive number size have been reported for just a smattering of isolates. Given this
significant gap in our knowledge, a systematic survey of HIV's physical properties would generate
valuable information that is unique in two respects. First, it would provide us with new
information on how physical properties vary from virus to virus. This would permit estimates of
the range and distribution of these properties for all isolates. Second, it would provide us with
information on the bulk behavior of the virus. This type of "macroscopic" data would complement
the more "microscopic" data that has been gleaned by studying IllV's molecular genetics. To
undertake this systematic survey, we must examine viruses from around the world and ask several
basic questions. For each physical property of DIIV, what is the entire spectrum of behavior? (For
example, what are the highest and lowest infectious fractions or the slowest and fastest rates of
decay?) Where do viruses from various geographic regions rank in this overall spectrum? How

does person-to-person transmissibility relate to this ranking? There are good reasons to believe
that today's more transmissible viruses will dominate the IIIV epidemic in the future. Answering