Stopping PCP Prophylaxis after Suppressing HIV

The old recommendations on the use of drugs to prevent new or recurrent AIDS-related opportunistic infections were based on a person's CD4 count. Patients began drugs to prevent pneumocystis carinii pneumonia (PCP) when their CD4 counts fell below 200 cells/mm3. Before the advent of combination antiviral therapy and advances in treatments for other OIs, PCP prophylaxis was the major means for prolonging life.

Now, there is mounting evidence that some persons with AIDS on successful anti-HIV therapy may be able to stop taking anti-OI drugs. A draft revision of the U.S. Public Health Service "Guidelines for the Prevention of Opportunistic Infections," issued on May 14, 1999, is urging consideration of stopping primary PCP prophylaxis in people whose CD4 count increases to above 200 cells/mm3 for at least three to six months.

Primary prophylaxis prevents a patient from ever getting a particular disease. Secondary prophylaxis is for preventing a return bout. The Guidelines judged it premature to make recommendations about secondary PCP prophylaxis. (The draft Guidelines also propose tentative criteria for discontinuing prophylaxis for certain other OIs: MAC -- ending primary prophylaxis when a patient's CD4 count is above 100 for three to six months; CMV -- secondary prophylaxis could be discontinued after someone's CD4 count is above 100 to 150 for three to six months and if only one eye is involved).

There is mounting evidence that patients with rebounding CD4 counts can tolerate discontinuation of PCP prophylaxis. In the April 17, 1999 issue of The Lancet, Gerrit Weverling and other members of the EuroSIDA study group published a report from a prospective observational study of 7,333 HIV-infected patients from 52 centers across Europe and Israel. The cohort included 378 persons who had discontinued PCP prophylaxis while on highly active antiretroviral therapy (HAART). Another 319 were still receiving primary prophylaxis (to prevent a first occurrence of PCP) while 59 were receiving secondary prophylaxis (to prevent a recurrence of PCP). The study characteristics are summarized in table 1.

Table 1: EuroSIDA Study
 Primary ProphylaxisSecondary Prophylaxis
Median CD4 cell counts prior to discontinuation of prophylaxis274270
Median CD4 cell count increase on HAART before discontinuing prophylaxis164174
Median lowest CD4 cell count ever documented12360
Median number of months with a CD4 cell count greater than 200 before discontinuing PCP prophylaxis85
Median number of months since stopping prophylaxis75

No cases of PCP were seen in any participant. Of particular interest over time will be the 59 who stopped secondary prophylaxis. Also, the fact that there were no reported cases of toxoplasmosis is significant because Bactrim and Dapsone, the most common antibiotics used to prevent PCP, also offer some protection against toxoplasmosis.

In another study published in the April 29, 1999 issue of the New England Journal of Medicine, Hansjakob Furrer and colleagues presented findings from the Swiss Cohort Study. Their article also dealt with the topic of the discontinuation of primary prophylaxis against PCP. Patients were eligible for this study if their CD4 cell counts had increased to at least 200 while receiving combination antiretroviral therapy, with these levels sustained for at least 3 months. Table 2 lists some of the study characteristics.

Table 2: Swiss Cohort Study
Total number of patients enrolled262
Number of patients who were positive for Toxoplasma antibodies at baseline121
Median CD4 cell count at study entry325
Median CD4 cell count nadir (lowest CD4 cell count ever recorded)110
Median HIV RNA level at study entry100
Number of patients with lowest CD4 cell count less than 5057
Number of patients on 3 or 4 drug combination therapy227
Number of patients on 2 drug combination therapy35
Number of patients with history of recurrent thrush, Candida esophagitis, weight loss, or fever113
Median follow-up in months11.3

Again, there were no cases of PCP pneumonia or toxoplasmic encephalitis.

Yet another report, published in the January 16, 1999 issue of The Lancet by Margaret Schneider and colleagues from the Utrecht University Hospital in the Netherlands, showed similar results (see table 3). No cases of PCP were diagnosed during the follow-up time. Two patients resumed prophylaxis due to low CD4 cell counts.

Table 3: Dutch Study
 Primary ProphylaxisSecondary Prophylaxis
Number of Patients (total 78)6216
Mean CD4 cell count at time of discontinuation353325
Lowest mean CD4 cell count during prophylaxis8553
Time with CD4 cell counts greater than 200 prior to discontinuation, in months7.07.3
Follow-up time in months147.8
Patients with undetectable viral load at time of discontinuation of prophylaxis 61 (78%)11(69%)

These three studies correspond with the mortality statistics that show a steady decline in the number of deaths from AIDS-related PCP. More individuals are living with an AIDS diagnosis then at any time before in the history of this epidemic. Yet the number of PCP cases has dramatically plummeted. This testifies to the success of HAART in preserving and restoring immune function.

Many questions remain to be answered when considering PCP prophylaxis. The foremost limitation of these observational studies is their brief follow-up time periods, the longest being 14 months. Also, definitive conclusions in this risky area are difficult to make because the data available do not come from controlled clinical trials. In a controlled clinical trial, the investigators would randomly assign volunteers to either stop prophylaxis or remain on it. After an appropriate period of time, the two groups would be compared for PCP incidence.

Certainly, the numbers of patients who discontinued their secondary prophylaxis is too few to be able to decide whether and at what CD4 cell count it might be safe to discontinue secondary prophylaxis. Finally, these studies give no guidance as to when someone who ceases prophylaxis for PCP might have to restart again. Will it be when CD4 cell counts dip below 200 again? Or should it be at a higher CD4 cell count if viral load rebounds?

Overall CD4 cell count is not the most precise way to evaluate the risks of stopping PCP prophylaxis. Assays that indicate specific aspects of restored immune response are now under investigation. These include lymphoproliferative assays that would directly measure cellular activation in response to PCP and other microbes. Still, the present reports are a further indication that the immune system can dramatically improve following suppression of HIV. For people with AIDS, they hold out the promise that antiviral therapy will render unnecessary the added burden of drugs to prevent opportunistic infections.

Back to the GMHC Treatment Issues April 1999 contents page.