Back in October, I predicted that the NIH's HIV treatment guidelines would eventually recommend starting treatment earlier than they do now (and Pharmalot.com wrote a response to this prediction analyzing how it might affect the market share of various drugs). Now there's more evidence that some of us may have paid a price for waiting until our CD4 counts fell below 350 before starting treatment. Turns out it's harder to get our CD4s back up to what's considered a normal level than it is for those that start treatment earlier (see this AIDSmeds story).
Same holds true for other immune reconstitution markers the Harvard researchers looked at, like the ratio of CD4 naive to CD4 memory cell counts, and the ratio of CD4 to CD8 cell counts (both ratios remained lower in the group that started therapy with less than 350 CD4 cells than in the groups that started in higher ranges). You can read the full study here (PDF file).