Long-term CD4 increase on HAART averages 250-350 cells, plateaus after 3-4 years

A series of four poster presentations at the 12th Retrovirus Conference looked at the immune recovery of patients staying on HAART for periods of between three and seven years and found that the average CD4 count increase achievable on HAART appears to be in the region of 350. They found median CD4 increases varying from 230 over 3.6 years in a Spanish study to 337 over 4.75 years in a French one.

The study with the longest follow-up period of all, which followed patients from the Swiss HIV cohort for seven years, found a CD4 count increase over that period of 349 in patients who had never interrupted therapy but only 153 in patients whose therapy had ever been interrupted, despite the fact that the median length of therapy interruption was only 32 days.

Three of the four studies also found that 11-14% of patients starting HAART at low CD4 counts never achieved a CD4 count over 200, while a fourth study with the shortest follow-up time (just under three years) found one in six patients did not a achieve a CD4 increase within that time of more than 100 cells.

In a discussion session, the poster presenters debated their conflicting conclusions as to whether CD4 cells continue to increase on HAART or reach a ‘plateau’ after 3-4 years; the consensus was that counts scarcely increase after this time except in patients who start with the very lowest CD4 counts.

An interesting chart on the Swiss HIV Cohort study poster illustrated that CD4 counts over time tend to converge. The largest CD4 increases over the seven years were in the people who started HAART at the lowest CD4 counts (under 100); conversely people starting HAART at the highest CD4 counts (over 700) actually had a slight fall in CD4 counts. Some investigators suggested this was evidence of a homeostasis mechanism in the immune system that adjusts T-cell counts to a pre-determined limit.

The four studies featuring long-term immune recovery were:

• The French APROCO study. This enrolled 1281 patients on to a protease-inhibitor-based regimen between 1997 and 1999. The poster looked at a subset of 438 patients who remained completely virologically suppressed, with no viral load reading over 500, over an average follow-up period of just under five years. It found a median CD4 rise of 337 cells in these patients and evidence of a ‘plateau effect’; after three years, the average CD4 count rise was 0.5 cells a month, statistically equivalent to zero. It found that 15% of patients starting HAART at a CD4 count under 100 never achieved a count over 200, even though remaining virally suppressed (Le Moing).
• The American ACTG 384 study. This study enrolled 980 people to compare the effect of six different NRTI/PI combinations. It had the shortest follow-up period (just under three years). The median CD4 count increase over this time was 252, with no evidence of a ‘plateau’ before three years, and 16 per cent of subjects had increases of less than 100 despite continued viral suppression. Women and younger people had slightly higher CD4 count increases. It found that patients with higher baseline viral loads had higher CD4 increases – the opposite to what the APRCO study found.
• The Spanish PISCIS study. This looked at 1542 patients treated mainly with PI-based regimens, 787 of whom remained virally suppressed over the median follow-up time of 3.6 years. This study looked at the patient group as a whole, not only at the fully-suppressed patients, and found a median CD4 count rise of 230 over this time, with eleven per cent of patients never achieving a CD4 count over 200. They found evidence of a ‘plateau’ after four years, except among patients who had started with a CD4 count under 100.
• The Swiss HIV Cohort Study. This found a median CD4 rise of 302 cells among 6497 patients followed-up for seven years. Forty-one per cent had CD4 increases of over 500 and 13 per cent of less than 200. Forty-two per cent remained virally suppressed for the seven years. As detailed above, the most striking finding was that having had a treatment break made a big difference to immune recovery.