The Honeymoon Continues
Data from the 23rd ACTG Meeting Document Reduction in Number of Opportunistic Infections in the Era of HAART
‘It’s a good thing’
Since the approval and wide-spread use of protease inhibitors in mid-1996, AIDS clinicians and primary care physicians across the country all insist they are seeing a marked decrease in AIDS-related opportunistic infections (OIs) in their HIV-positive patients. Until now, this was all wonderful yet anecdotal news. At the 23rd AIDS Clinical Trials Group (ACTG) Meeting in Washington, D.C. (July 19-23, 1997), a number of investigators presented epidemiologic data which clearly demonstrate that the rate of most OIs has indeed fallen.
The most striking presentation, which one would have missed unless one happened to stumble into the Protozoa Pathogen Study Group symposium, came from Tulane University’s Rebecca Clark. Clark presented epidemiologic data from the CDC sponsored Adult Spectrum of Disease Cohort (ASDC) from December 1994 to July 1997. Baseline characteristics were used to group 1,181 patients before 1996 and 1,248 patients after 1996. A quick scan of the baseline characteristics revealed three important statistically significant differences between the groups: 1) The percentage of women in the latter group increased from 15.6% to 19.5% (p=0.01); 2) The percentage of gay/bisexual individuals in the latter group decreased from 49.7% to 43.1% (p<0.01); and 3) The latter group improved immunologically (53.5% of the individuals before 1996 had CD4 counts <50 cells compared to 43.5% after 1996 (p<0.01)). Although these data extend only through the first six months of 1997, it is presumed that the the trend will continue throughout the end of the year. When comparing the rate of the most common OIs, including Pneumocystis carinii pneumonia (PCP), cytomegalovirus (CMV) Mycobacterium avium complex (MAC), esophageal candidiasis, and wasting from 1994, 1995 and 1996 with those from the first six months of 1997, the decrease is staggering.
In comparing the 18 months before and after January 1996, however, the data reveal that only 5 of the 11 OIs have decreased significantly: MAC, CMV, KS, wasting, PCP. Decreases in the incidence of cryptococcal meningitis, toxoplasmosis, cryptosporidiosis, and esophageal candidiasis, by contrast, were not significant (although the reductions in meningitis and toxoplasmosis approached significance). Furthermore, the marked and visible decrease in opportunistic infections in Clark’s study did not occur until the calendar year 1997 -- even though 3 of the 4 protease inhibitors were approved and available by mid-1996. Is it enough to say that these decreases are solely due to the antiviral effect of triple combination therapy? Might there not be other factors involved, such as increased use MAC prophylaxis, steroid usage or dietary changes for the prevention of wasting, or a change in sexual behavior in order to avoid KSHV and thus Kaposi’s?
A preliminary data analysis of the OI events in ACTG 320 was presented by USC’s Judith Currier, an excellent young investigator who’s taking the ACTG by storm. ACTG 320 was Merck’s indinavir clinical endpoint study which randomized 1,156 patients with CD4 counts <200 and at a minimum of six months’ prior AZT therapy to add either 3TC (n=579) or to add both 3TC and indinavir (n=577). To no one’s surprise, ACTG 320 was stopped early because of statistically significant differences in deaths and AIDS events favoring the protease inhibitor-containing group. There were 60 OIs in the group that added one drug and 31 OIs in the group that added both 3TC and indinavir. A detailed analysis of the clinical events in ACTG 320 is currently underway (as well as a possible comparative analysis with ACTG 175). Nonetheless, it is probably safe to say that the incidence of OIs in the triple therapy arm of ACTG 320 is the lowest ever documented in a study of this size with this patient population. It will be interesting to see whether or not some of these OIs (i.e., PCP and MAC) developed in individuals on prophylaxis and also at what CD4 count. That is, Did increases in CD4 cell counts protect these individuals from new opportunistic infections?
CD4 cell increases and an "undetectable" viral load are one thing, but this decrease in life-threatening OIs is the proof of the pudding and arguably the reason we are finally seeing a substantial decline in AIDS deaths. Numerous studies have documented the deleterious effect OIs have on HIV-infected persons: the inflammatory cascade where one OI seems to lead directly to another; the independent predictability of at least six OIs to shorter survival; and the well-documented viral load surges which accompany many OIs. So, in the ineradicable mantra of Martha Stewart, this reduction in OIs is "a good thing."
Nevertheless, investigators at the ACTG meeting expressed caution that we are in a "honeymoon phase." Clinical follow-up of ACTG 320 was, after all, less than one year. And widespread use of protease inhibitors has just recently crossed the one year mark. There are many who fear that in 6 to 18 months we might see a substantial number of patients fail their protease inhibitor regimens (because of resistance and noncompliance) and experience anew a marked increase in the incidence of opportunistic infections.
If this is indeed to be the case, only effective OI prophylactic regimens, (TMP-SMX for PCP and clarithromycin or azithromycin for MAC) can insure patients with viral break-through and faltering CD4 cell counts protection from the opportunistic infection menace. Even then, prevention of CMV will remain problematic. Roche’s outrageously expensive oral ganciclovir (the closest thing we have to a prophylactic against CMV -- whatever happened to lobucavir?) is only truly effective in CMV DNA-PCR-negative patients (that is, those at the lowest risk of developing CMV disease). What are we to do with those high-risk patients with high levels of CMV DNA in their blood? Don’t look for any help from Basel. Roche execs are said to be reconsidering the potential market for its oral ganciclovir prodrug (RS-79070) now that the incidence of CMV disease appears to have fallen so. But such a short-sighted business decision would be, well, myopic. Even if this business is all about the bottom line, slowing or discontinuing the development of RS-79070 is (in as little as a year or two from now) very likely to be exposed as strategic miscalculation. Woe to s/he who wields that red pen.
The Opportunistic Infections of ACTG 320
*Statistically significant (for CMV, p=0.055)