Inflammation is increasingly recognized as an immunological double-edged sword: it contributes importantly to the response to infection, but can also cause serious collateral damage to the body—particularly when persistent—and has been implicated as potentially contributing to multiple conditions, including heart disease and cancers. Researchers have found that the anti-inflammatory effects of statin drugs contribute to their beneficial effects against cardiovascular disease, but because statins also reduce cholesterol it has not been possible to disentangle the relative importance of these two potential mechanisms of action.
To directly address the role of inflammation in cardiovascular disease, the pharmaceutical company Novartis sponsored a 10,061-person randomized trial of their antibody canakinumab, which is designed to reduce inflammation by blocking the cytokine IL-β (it is already FDA-approved for the treatment of a number of other conditions). The trial, named the Canakinumab Antiinflammatory Thrombosis Outcome Study (CANTOS), recruited stable HIV-negative individuals who had previously experienced a heart attack and had levels of the inflammatory biomarker high-sensitivity C-reactive protein (hsCRP) of 2 mg or more per liter. The average age of participants was 61 years old. Three different doses of canakinumab were studied: 50 mg, 150 mg, or 300 mg, administered subcutaneously every three months. The primary endpoint was nonfatal heart attack, nonfatal stroke, or cardiovascular death.
The main results were published in the New England Journal of Medicine on August 27, 2017, drawing considerable media attention. After an average of 3.7 years of follow up, the results showed a slight but statistically significant reduction in the incidence of primary endpoints in the 150 mg dose group: there were 3.86 events per 100 person-years compared to 4.50 events per 100 person-years in the placebo group (an approximately 15% reduction in risk). There was a similar reduction in the 300 mg group (3.90 events per 100 person-years) but it did not meet the criteria for statistical significance. No significant effect was observed in recipients of the 50 mg dose. The most common primary endpoints were heart attacks.
In a separate paper published in The Lancet, another interesting possible benefit was reported against cancer; the analysis was not pre-planned so is considered exploratory, but there was a statistical trend toward a reduction in cancer mortality when recipients of all canakinumab doses were compared to the placebo group. The result achieved statistical significance in recipients of the 300 mg dose, for whom the risk of cancer mortality was roughly halved compared to placebo. The incidence of lung cancer was also significantly reduced in the 150 mg and 300 mg dose groups, and there was a reduction in lung cancer-related mortality that reached statistical significance in the latter group.
Consistent with the clinical outcomes, canakinumab was associated with dose-dependent reductions in the levels of the inflammatory biomarkers hsCRP (26–41%) and interleukin-6 (25–43%).
There were distinct downsides to the therapy: fatal infections or sepsis were significantly more frequent among canakinumab recipients (the incidence rates were 0.31 vs. 0.18 events per 100 person-years in the canakinumab and placebo groups, respectively), and this meant that all-cause mortality was similar between the treatment and placebo arms. Neutropenia and thrombocytopenia were also significantly more frequent in the canakinumab groups.
In an editorial commentary in the New England Journal of Medicine Robert Harrington notes that uncertainties about the risk/benefit of canakinumab that will need to be resolved before it can be considered for routine use against cardiovascular disease. Harrington also highlights the current exorbitant price of the antibody: approximately $200,000 per year in the United States. More optimistically, he emphasizes the importance of the CANTOS trial in confirming that targeting inflammation can produce clinical benefits in cardiovascular disease. Other trials are addressing whether cheaper alternative anti-inflammatories can produce similar or better results, with one example being an evaluation of low-dose methotrexate sponsored by the National Heart, Lung, and Blood Institute.
Implications for HIV?
Associations between higher levels of inflammatory biomarkers and increased risk of morbidity and mortality (including events resulting from cardiovascular disease and cancers) have been consistently reported in HIV-positive people, most notably in analyses of the large SMART and ESPRIT trials. A paper just published in the Journal of AIDS reports that levels of the biomarkers IL-6 and D-dimer were also significantly associated with potentially life-threatening grade 4 events that were not attributable to AIDS, cardiovascular disease, or non-AIDS cancers in these study populations. These results prompt the question of whether canakinumab or other anti-inflammatories might have salutary clinical effects in HIV-positive people.
The researcher Priscilla Hsue at UCSF has conducted a small 10-person pilot study of a single 150 mg canakinumab dose in HIV-positive people on ART – results were described earlier this year at CROI (a webcast of the presentation is available online). Participants were required to be over 40 years old with known cardiovascular disease or at least one risk factor.
Eight weeks after the single dose, there were dramatic declines in levels of hsCRP (41%) and IL-6 (30%), as well as evidence of reduced arterial inflammation. Adverse events included a 28% drop in absolute neutrophil count measured at two and three weeks after canakinumab administration, which recovered to normal by week four. There was also a single case of shingles that may have been related to the study drug, although Hsue noted that no cases were observed in the CANTOS trial.
Hsue is now conducting a larger randomized, placebo-controlled extension of the trial that will administer two doses of canakinumab, at baseline and week 12, and follow participants for 36 weeks. The study aims to recruit 100 HIV-positive individuals and has the same entry criteria as the pilot: over 40 years of age with known cardiovascular disease or at least one risk factor. Results should help discern if canakinumab might have potential in HIV, and it will be interesting to find out if the effects on hsCRP and IL-6—which to my knowledge dwarf the reductions seen with any other interventions—are sustained in the absence of ongoing repeat dosing.
In terms of other research involving potential anti-inflammatory approaches, the most significant is the large REPRIEVE trial, which is evaluating the clinical effects of pitavastatin in 6,500 HIV-positive individuals (aged 40-75) on ART. Substudies will investigate the relationship between effects on inflammatory biomarkers and outcomes. An AIDS Clinical Trials Group (ACTG) study of low-dose methotrexate in 176 HIV-positive ART-treated participants over 40 years of age has been completed and results are pending.
Looming over the research into therapeutic approaches for reducing inflammation in HIV-positive people on ART are unresolved questions about the exact causes of the problem. Studies have been consistent in showing a relationship between the magnitude of persistent inflammation and timing of ART initiation—the later in the course of infection that ART is started, the higher the levels of inflammatory biomarkers and vice versa. But even in people who start ART very early, inflammatory biomarkers typically don’t drop all the way down to levels that are normal for healthy HIV-negative people, and the exact causes are still unclear.
Some research has posited that the persistence of HIV contributes, but a large ACTG study published earlier this year found no significant relationship between various measures of HIV persistence (e.g. HIV DNA, HIV RNA) and inflammatory biomarkers in individuals on long-term ART. Rather, the levels of inflammation remained significantly associated with pre-ART measures, which led the researchers to conclude that “immune events that occurred before treatment initiation had long-lasting effects despite sustained control of the virus.”
While the study could not answer the question of exactly what the immune events were, the authors write: “Mechanisms consistent with our observations are those that predominate during untreated infection but can exert a long-lasting effect, such as fibrotic damage to lymphatic tissue, increased intestinal permeability leading to elevated microbial translocation, or persistent co-infections in the setting of compromised immune surveillance.” There has been some research into approaches that might address these potential causes, particularly candidates predicted to affect microbial translocation such as sevelamer, mesalamine, rifaximin and various probiotics, but results have been mixed, with little sign of significant anti-inflammatory effects. A trial of the antifibrotic drug losartan is currently ongoing.
Another possible contributor to persistent inflammation might be immune system perturbations such as loss of naïve T cells and inversion of the CD4/CD8 ratio, which tend not to fully normalize even in early-treated individuals. There was some intriguing evidence that the cytokine IL-7, which promoted reconstitution of naïve T cells in people on ART, was associated with short-term declines in inflammatory biomarkers, but since the manufacturer went bankrupt several years ago this candidate immune-based therapy appears to be in limbo.
Additional studies on the causative mechanisms of persistent inflammation on ART are clearly needed, in order to better define targets for therapeutic interventions.