Persistently Low CD4 Counts in Patients with Suppressed HIV RNA
Among patients who take antiretroviral therapy and achieve suppression of HIV RNA levels, most have a substantial increase in their CD4 cell count. Typically, patients have a brisk increase in CD4 cells in the first 3 to 6 months after starting antiretroviral therapy, predominantly due to a release of memory CD4 cells trapped within lymphoid tissue. In the second phase of CD4 recovery, there is a gradual increase in CD4 counts that continues for 3 to 6 years; this phase involves both naive CD4 cells (from the thymus) and memory CD4 cells. Approximately one-third of patients who maintain continuous suppression of HIV do not recover their CD4 cell count to a level above 500 cells/mm3 after 5 years. A smaller proportion of patients (less than 10%) fail to recover their CD4 count at a level greater than 200 cells/mm3 despite virologic suppression. This is often referred to as a "discordant" or " immuno-virological discordant" response (good virologic response and poor immunologic recovery). This discordant response is associated with increased risk for developing an opportunistic infection and increased progression to AIDS or death,[,,,] but the risk of developing new AIDS-defining event declines substantially after the first 6 months of virologic suppression. "Discordant responses" can also refer to good immunologic responses despite incomplete virologic suppression, but the following discussion will address only discordant responses with poor immunologic recovery.
Immunologic Recovery Based on Specific Antiretroviral Regimens
Most patients who achieve sustained virologic suppression with one of the recommended modern antiretroviral therapy regimens have good CD4 count recovery. Existing data suggest that in antiretroviral-naive patients, efavirenz (Sustiva) produces lower CD4 cell count recovery than with boosted-protease inhibitors, maraviroc (Selzentry), and raltegravir (Isentress). In ACTG 5142, patients receiving lopinavir-ritonavir (Kaletra) plus two nucleoside reverse transcriptase inhibitors (NRTIs) had greater CD4 cell count increases at 96 weeks than patients taking efavirenz plus two NRTIs (287 versus 230 cells/mm3). A meta-analysis reported that mean CD4 cell count increases at week 48 were better with a ritonavir-boosted protease inhibitor regimen (+200 cells/mm3) than a non-boosted protease inhibitor regimen (+179 cells/mm3) or a nonnucleoside reverse transcriptase inhibitor (NNRTI)-based regimen (+173 cells/mm3)(Figure 1). This meta-analysis, however, had multiple confounding factors, and the NNRTI-based regimens included data for nevirapine. Analysis of the Swiss HIV Cohort study also showed a trend for better CD4 responses with boosted-protease inhibitor regimens (Figure 2). Similar CD4 cell count recovery occurs with most, but not all, NRTIs. The combination of tenofovir (Viread) and didanosine (Videx) has been associated with lower CD4 cell count responses, particularly when the didanosine dose exceeds 4.1 mg/kg (Figure 3). In addition, zidovudine-based regimens may have poorer CD4 count responses, presumably because of the marrow suppressive effect of zidovudine.
Causes of Discordant Responses with Poor Immunologic Recovery
When patients have poor CD4 cell count responses in the setting of sustained virologic suppression, several potential reversible causes should be considered, including receipt of marrow-suppressive medications and infiltrative bone marrow processes. Common marrow suppressive drugs used in HIV-infected patients include zidovudine (Retrovir) and zidovudine containing fixed combination pills (Combivir and Trizivir), trimethoprim-sulfamethoxazole (Bactrim, Septra), interferon and peginterferon (Pegasys or PegIntron), pyrimethamine, sulfadiazine, ganciclovir (Cytovene), valganciclovir (Valcyte), and etoposide. Medication-related marrow suppression is more likely to occur when a combination of marrow suppressive agents are used, such as zidovudine plus trimethoprim-sulfamethoxazole. Marrow infiltrative processes, such as lymphoma and disseminated histoplasmosis, should also be considered. Only after excluding reversible causes of poor immunologic should the patient be considered to have a true discordant response with poor immunologic recovery. Factors identified with poor immunologic recovery include low baseline CD4 cell count,[,,] older age, and possibly co-infection with hepatitis C virus.
Interleukin-2 for Persistently Low CD4 Cell Counts
Several studies clearly established that interleukin-2 (Aldesleukin) given with antiretroviral therapy causes substantially greater increases in CD4 cell counts than antiretroviral therapy alone. In contrast, interleukin-2 given without antiretroviral therapy generates inferior CD4 count responses when compared with antiretroviral therapy alone. These studies, however, did not evaluate the clinical impact of interleukin-2 therapy. Patients who have a discordant response and CD4 counts persistently less than 200 cells/mm3 were of particular interest for potential investigational use of interleukin-2. Specifically, the hope was that patients with sustained virologic suppression but poor immunologic recovery could substantially increase their CD4 cell count using interleukin-2, thereby reducing their risk of opportunistic infection and death. To determine whether the addition of interleukin-2 to antiretroviral therapy reduced the risk of opportunistic diseases or death, the NIH sponsored two large phase 3, randomized, international trials: (1) Evaluation of Subcutaneous Proleukin in a Randomized International Trial (ESPRIT) and (2) Subcutaneous Recombinant Human IL-2 in HIV-infected Patients with Low CD4 Counts under Active Antiretroviral Therapy (SILCAAT). In ESPRIT, 4,111 patients with CD4 counts greater than 350 cells/mm3 were randomized to receive interleukin-2 (7.5 MIU twice daily for 5 consecutive days every 8 weeks for at least 6 months) plus antiretroviral therapy or antiretroviral therapy alone; patients had an average follow-up of 7 years. Although patients who received interleukin-2 and antiretroviral therapy had an average increase in CD4 count that was 159 cells/mm3 greater than the group receiving antiretroviral therapy alone, there were no differences in clinical outcomes (Figure 4).  Investigators in SILCAAT randomized 1,695 patients with a CD4 cell count between 50 and 299 cells/mm3 to receive interleukin-2 plus antiretroviral therapy or antiretroviral therapy alone, with follow-up of approximately 7 years. Interleukin-2 was given at a dose of 4.5 MIU twice daily for 5 consecutive days every 8 weeks for 49 weeks. Patients who received interleukin-2 and antiretroviral therapy had an average CD4 count 53 cells/mm3 greater than those who received antiretroviral therapy alone, but no differences in clinical outcomes were observed (Figure 5). The reason for the lack of clinical benefit despite increased CD4 counts remains unclear.
Recommendations for Patients with Persistently Low CD4 Cell Counts
For persons who have sustained virologic suppression for at least 2 years but have CD4 counts consistently less than 200 cells/mm3, the following approach is recommended. First, make certain the patient is receiving appropriate prophylaxis for opportunistic infections. Second, examine the patient's medication list for medications that can suppress bone marrow. In patients taking a potentially marrow suppressive drug, change the medication to a non-marrow suppressive drug, if possible. For example, consider switching from a zidovudine-containing regimen to a regimen that does not contain zidovudine. Third, evaluate the patient for clinical manifestations, such as systemic symptoms or pancytopenia, which suggest a marrow infiltrative process. Fourth, continue antiretroviral therapy, even if the patient has not had a good CD4 cell count response. Multiple studies have shown that achieving a durable virologic response translates into clinical benefit independent of CD4 count.There are no "switch" data that support a change from one suppressive regimen to another, although the combination of tenofovir plus didanosine (especially at full dose) should be avoided, and there may be potential benefit of switching to a non-zidovudine-containing regimen. There is evidence, discussed above, that regimens containing ritonavir-boosted protease inhibitors, maraviroc, and raltegravir result in greater CD4 cell count responses than efavirenz-based regimens, although there is currently little evidence supporting modification or intensification of efavirenz-based regimens in patients with discordant CD4 responses. The intensification approach is currently being evaluated in clinical trials. Finally, existing data do not support the use of interleukin-2 in this setting. Although preliminary data with other investigational agents, such as interleukin-7, have demonstrated an increase in CD4 cell counts, there are no clinical data to support the use of such therapies in clinical practice.