Abstract
Article abstract Introduction of highly active antiretroviral therapy (HAART) has been associated with many changes in the complications of human immunodeficiency virus (HIV) infection. A cohort of 25 HIV patients with progressive multifocal leukoencephalopathy (PML) treated with HAART experienced a median survival of >46 weeks. This is an improvement in prognosis compared with recent historic experience and correlated with HIV RNA viral load reductions. We conclude that current HIV therapy is important in improving the outlook of PML in the setting of HIV.
Progressive multifocal leukoencephalopathy (PML) is one of the most devastating opportunistic neurologic infections complicating the lives of immunodeficient patients. PML is caused by the JC virus, a commonly acquired and generally innocuous polyoma virus. Although PML used to be a rare condition, seen primarily in patients with malignancies, it became a frequent complication of HIV-infected patients during the 1980s (approximately 4%).1 A recently reported blinded treatment trial for patients with PML with cytarabine (AIDS Clinical Trial Group [ACTG] study 243) emphasized the rapidly progressive nature of PML in this population.2 This study, conducted before the general availability of protease inhibitors, demonstrated no survival advantage of cytarabine over the then current best antiretroviral therapy. The median survival was less than 11 weeks from the time of initiation of treatment. No other controlled trials have demonstrated therapeutic efficacy against this opportunistic infection.
Human immunodeficiency virus (HIV) therapy has been transformed by the introduction of the first protease inhibitor therapy in late 1995. In the subsequent year, highly active antiretroviral therapy (HAART)—with its use of multiple antiretroviral drugs including various combinations of nucleoside reverse transcriptase (RT) inhibitors, non-nucleoside RT inhibitors, and protease inhibitor (PI) antiretrovirals—became the standard practice for HIV treatment in the United States. The HAART regimen is guided by plasma viral load measurements, with the goal of achieving nondetectable HIV RNA viral loads in the patient’s plasma. These practices have resulted in a dramatic improvement of life expectancy in HIV patients.3 Recognizing the unknown impact of this therapeutic change on the course of PML, we undertook an observational study of the natural history of PML in the era of HAART. The purpose was to obtain information critical to recommendations of therapy for PML if warranted, and to provide a basis for development of treatment trials for this rapidly fatal neurologic illness.
Methods.
The diagnosis of PML was made by experienced clinicians at 10 institutions, the majority of whom had accrued the controlled treatment population in ACTG 243. For inclusion in the current study, diagnosis of PML was made either by a brain biopsy pathologically compatible with PML or by a confluence of findings consisting of identification of JC virus DNA in the CSF by PCR, a progressive neurologic deficit, and a MR brain scan demonstrating noncontrast enhancing white matter disease. The majority of subjects (15) were taking HAART therapy at enrollment. The investigators all recommended use of HAART therapy guided by viral load measurements. Data were collected with the informed consent of subjects. Clinically indicated measurements of HIV RNA viral load measurements were collected on a periodic basis. The primary objective was to determine duration of survival in this group. In the case of deaths, the date and presumed cause of death was identified.
The survival distribution (probability of patient survival beyond a given point in time) was estimated using the Kaplan-Meier method. Survival distributions between subject groups were estimated by the method of Kaplan and Meier and compared by the log-rank test.
Results.
The rate of accrual of the 25 subjects was approximately 2.0 patients per month over the year of observation. Of the 25, 17 had a compatible brain biopsy, with the remainder demonstrating JC DNA in the CSF with compatible clinical and imaging lesions. The median age of the population was 39.5 years (range, 26 to 56). The majority (92%) of study subjects were men, and 56% were of Caucasian, non-Hispanic race/ethnicity. The baseline median CD4 lymphocyte count was 104 cells (range, 4 to 1,030). Median baseline viral load was 3,362 RNA copies per milliliter (range undetectable to 1.5 million), with eight subjects having less than 500 copies/mL at baseline.
As is typical for advanced HIV patients, these subjects took multiple prophylactic medications, including sulfa drugs, antifungals, and macrolide antibiotics. In addition, four subjects received therapeutic agents directed against PML (three prescribed interferon-α and one cidofovir). HAART therapy including protease inhibitor antivirals was employed at baseline in 15 subjects. Mean duration of HAART therapy for those with viral load <500 copies of RNA/mL was 9.3 months (range, 2 to 17); the mean duration of HAART for those with viral load >500 was 5.25 months (range, 1 to 17).
The survival experience is plotted in figure 1. Median survival for these subjects was 46.4 weeks (range, 0 to 51 weeks). This is contrasted with the median patient survival in ACTG 243 of 11 weeks from study entry.
Figure 1. Overall patient survival from study entry. AIDS Clinical Trial Group (ACTG 243) patient survival is superimposed for comparison. The median survival time (time to death in weeks since study entry) was 10.6 weeks among ACTG 243 patients; Neurologic AIDS Research Consortium (NARC 001) patients were observed to be living longer, with a median survival time of 46.4 weeks. Both estimates are based on the method of Kaplan and Meier.
HIV RNA viral load, used as a surrogate marker for the success of HAART, was chosen as a means of investigating the impact of HAART on survival. Subjects were stratified according to whether they had attained HIV viral load suppression below 500 copies/mL. At baseline, viral load suppression did not appear to be predictive of subsequent survival (log-rank p = 0.62), whereas the most recent viral load was highly predictive (p = 0.03). Subjects responding well to HAART with HIV viral loads below 500 copies/mL had significantly longer survival than those with poorer viral control reflected by viral load >500 copies (figure 2).
Figure 2. Patient survival from study entry, according to viral load suppression (most recent measurement).
The impact of the patients’ relative immune competence on survival was investigated by stratification by CD4 lymphocyte count at baseline. Subjects with CD4 counts less than 100 cells/mL (roughly half of the study cohort) had significantly lower survival compared with patients with CD4 counts over 100 (log-rank p = 0.003; figure 3). Stratification of the patient population according to higher CD4 cutoff points (e.g., higher versus lower than 200 cells/mL, or higher versus lower than 300 cells per mL) was not predictive of subsequent survival. Because CD4 was not frequently used in follow-up of these subjects, we have insufficient data to determine if the change in CD4 correlates with survival.
Figure 3. Patient survival according to CD4 count at baseline. Subjects were stratified according to CD4 count less than versus larger than 100 cells/mL.
Discussion.
This cohort suggests that HAART therapy, with its accompanying HIV RNA viral load reductions and presumed CD4 lymphocyte count elevation, promotes significant gains in survival for HIV patients with PML. Until recently, the prognosis of PML was deemed to be so dismal that efforts to treat these patients outside of research programs were often forsaken. This experience suggests that maximally effective HIV therapy should be recommended.
This series is the largest reported experience addressing PML survival in the HAART era. Prior reports of anecdotal experiences with individual or smaller groups of patients showed improvement and long term survival associated with HAART therapy.4-9 However, because of the known natural history of this infection, in which a minority of patients stabilize or remit, a larger experience is critical to substantiate a change in prognosis for this infection.
PML remains a considerable problem in the HAART era. The rate of accrual (2.0 subjects per month) experienced for this cohort is comparable to the controlled trial (ACTG 243) study group experience (2.3 per month). However, it is impossible to evaluate the current incidence of PML based on the data available from this study. It is important to note that this complication occurred in patients on HAART (15/25) and even in subjects with substantial control of viral load. This continued occurrence of PML may relate to the fact that it can present at relatively higher CD4 counts than some other opportunistic complications, suggesting that even patients with partially recovered immune competence are still at risk for this complication.1 However, the patient’s intrinsic immune response appears important in the course of PML. These results support the recent finding reported by Berger et al.10 of improved prognosis in subjects encountering the disease with higher CD4 counts. The application of HAART may achieve both increased CD4 lymphocyte counts and suppressed HIV viral load, the factors associated in this cohort with improved survival. It is notable that the survival of subjects not responding to HAART closely replicated the survival observed in the pre-HAART era.
The fact that PML occurs in some patients with undetectable HIV viral loads treated with HAART and that patients continue to die from this complication supports an ongoing need to address more direct modes of therapy for PML. Ongoing research seeking to evaluate new therapies including topoisomerase inhibitors and nucleotide antivirals are warranted in light of the occurrence and mortality of PML.
Acknowledgments
Supported by P05 R01 NS32228-04, National Institute of Neurological Disorders and Stroke (National Institutes of Health), to D.B.C.
Acknowledgment
Assistance of the Neurologic AIDS Research Consortium Performance and Safety Monitoring Committee—consisting of Burk Jubelt, MD, Bruce Barton, PhD, John H. Noseworthy, MD, Leroy Sharer, MD, A.P. Kerza-Kwiatecki, PhD, and F.J. Brinley, MD—is gratefully acknowledged. The authors thank all the patients and their families for participation in this study.
Footnotes
-
See also page 440
- Received October 6, 1998.
- Accepted November 21, 1998.
Letters: Rapid online correspondence
REQUIREMENTS
You must ensure that your Disclosures have been updated within the previous six months. Please go to our Submission Site to add or update your Disclosure information.
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
