latent TB in HIV: INH plus rifapentine for 1 month

A recent article found that one month of rifapentine plus isoniazid (INH) daily was noninferior to a nine-month INH course in patients with HIV and either latent tuberculosis infection (LTBI) or living in an area with high TB prevalence (see tb ltbi hiv inh rifapentine 1 month nejm2019 in dropbox, or DOI: 10.1056/NEJMoa1806808) 

  

Details: 

-- 3000 HIV-infected people living in areas of high TB prevalence (at least 60 cases per 100,000 population) or having evidence of LTBI. From 45 sites in 10 countries. 

-- Randomized to open label regimen of daily rifapentine (300mg a day for weight <35kg, 450mg per day for weight 35-45 kg, and 600mg a day for weight >45kg) plus INH 300mg plus pyridoxine for one month; vs INH plus pyridoxine for 9 months 

-- 54% women, median age 35, 52% from Africa/8% Asia/24% South America/16% North America, 66% black non-Hispanic/24% Hispanic/8% Asian or Pacific Islander, BMI 23.5, median CD4 count 470 (87% with CD4 count >250), 43% on efavirenz-based ART/7% nevirapine-based ART/50% on none, HIV viral load undetectable in 77% of those on therapy 

-- PPD positive in 21%, negative in 68% (ie, mostly just from high prevalence country); IGRA positive in 2%, negative in <1% 

-- 97% lived in areas of high TB prevalence 

-- primary endpoint was 1st diagnosis of TB, death from TB or death from an unknown cause 

-- patients followed 3.3 years 

  

Results: 

-- primary endpoint was found in 32 of 1488 patients (2%) in the 1-month group vs 33 of 1498 (2%) in the 9-month group: incidence of 0.65 per 100 person-years vs 0.67 per 100 person-years, finding noninferiority of the 1-month therapy 

    -- active TB accounted for 29 events in the 1-month group (91%), including 18 confirmed cases and 11 probable ones; vs 26 in the 9-month group (79%), with 14 confirmed cases, 10 probable and 2 confirmed TB-related deaths.  

    -- There were a total 6 TB-related deaths, 3 in each group and 1 attributable to immune reconstitution inflammatory syndrome  (no further specifics in article or supplementary material)

--in a per-protocol analysis (ie, including only patients adherent to med-taking/study guidelines): incidence rates were quite similar: 0.64 and 0.61 per 100 person-years in the 1-month and 9-month groups 

--subgroups (though all differences were not statistically significant): 

    -- LTBI subgroup had higher rates of primary endpoint (0.90 and 0.97 per 100 person-yrs, in the 1- and 9-month groups), vs those just living in high prevalence areas (0.58 and 0.59) 

    -- patients on antiretroviral therapy (ART) at entry: 0.55 and 0.63 per 100 person-yrs, in the 1- and 9-month groups; vs not on ART, 0.75 and 0.72) 

    -- by CD4 count: those <250 had primary endpoint in 1.93 and 1.28 per 100 person-yrs, in the 1- and 9-month groups; vs those with CD4 >250 at 1.28 and 0.59 (ie, seemed to be much better response in those with CD4 >250, though pretty few patients had the lower CD4 level, and there actually were too few to determine inferiority or non-inferiority along with very large confidence intervals) 

-- treatment completion was 97% in the 1-month group and 90% in the 9-month group (p<0.001) 

-- only one patient in each group developed TB with a rifampin-resistant strain; 2 patients in the 1-month group and 1 in the 9-month group developed TB with an INH-resistant strain  

-- serious adverse effects occurred in 6% of the patients in the 1-month group and 7% in the 9-month group (p=0.07); 1% of patients had meds withheld for >7days in the 1-month group vs 2% in the 9-month group, with OR 2.09 (1.32-3.33). neuropathy occurred in 2% of patients in 1-month group vs 3% in the 9-month group, and increased liver enzymes in 2% vs 3%, though neutopenia was more common in the 1-month group at 2% vs 1% 

  

Commentary 

-- tuberculosis is the leading killer of patients with HIV: about 1000 people with HIV die each day from tuberculosis (and, overall, estimates are that between ¼ and 1/3 of the world’s population is thought to have asymptomatic LTBI) 

-- in 2017 fewer than 1 million HIV-infected patients received preventive therapy, though there was an estimated 30 million who were eligible. And LTBI therapy pretty clearly decreases risk of death: in HIV-positive patients, this is true independent of whether they are on ART or not. 

-- the reason for the above study is that adherence rates to long-term INH in the community is low, around 50 to 70% for those completing the requisite 6 to 9 months

-- 3 months of weekly INH/rifapentine in a supervised setting (directly observed therapy) has completion rates in the 87-96% range in research settings and 82% in clinical practice 

-- overall, my general bias in using INH-based therapies is to add pyridoxine. Significant potential benefit (decreasing neuropathy), and no evident harm or significant expense 

-- it was important to note that there were low rates of drug-resistant tuberculosis in those who did develop active tuberculosis 

-- the biggest obstacle in resource-rich countries to the above 1-month course is drug-drug interactions with the HIV meds. http://gmodestmedblogs.blogspot.com/2018/07/updated-ltbi-treatment-with-weekly-meds.html has a recent blog on the CDC’s recommendation for the rifapentine/INH 3-month weekly therapy, including a link to a website to check rifapentine drug-drug interactions (in general rifapentine should not be administered with protease inhibitors; no problem with efavirenz but should not be given with nevirapine, etravirine, doravirine, rilpivirine; it should not be given with the INSTIs (but can be used with raltegravir 400mg bid; and there is no problem with giving the rifapentine/INH once weekly with raltegravir); and should not be given with TAF (tenovofir alafenamide), since rifapentine may decrease TAF concentrations. it should also not be given with maraviroc. Most of these contraindications were because of drug interactions that lowered the levels of HIV meds 

    -- the HIV drug interaction website i use most is the one from Unversity of Liverpool (https://www.hiv-druginteractions.org/checker ) which actually gives the same relatively bland "potential interaction" label for dolutegravir (same as given for efavirenz, nevirapine, and raltegravir, as well as maraviroc).  though should not be given with any TAF-containing regimen (eg biktarvy: bictegravir/TAF/FTC)

-- an evident issue raised by this study is the likely effectiveness of this 1-month therapy in patients who do not have HIV. I personally have been prescribing mostly the 12-week course of weekly rifapentine/INH, though there are health center resources tied up with that (one of our nurses or medical assistants calls the patient every week to remind them). And the other regimen I use has been rifampin for 4 months (shorter than INH and less toxic). Both without a problem. But in HIV patients, given the limiting drug-drug interactions, I still prescribe the 9-month INH course (though I did have one patient landing in the ICU for weeks from near-fatal INH hepatitis) , unless they are still on atripla (efavirenz with TDF/FTC) where rifapentine/rifampin based regimens are okay

-- limitations of the study: not include people <13 yo, pregnant or breast-feeding women 

so, there seem to be significant benefits from the shorter 1-month therapy, specifically patients were more likely to complete therapy and had fewer severe adverse events; and the short-course therapy is not inferior in terms of developing TB endpoints. Also, a shorter course of therapy is more efficient in terms of utilization of health care resources. At this point, given our level of knowledge about drug-drug interactions, for patients in resource-rich countries, it would probably mean using the older (but still quite effective) efavirenz/TDF/FTC combo, at least for the duration of TB therapy. And, based on inadequate data in this study, I would be hesitant to use the 1-month course in those with CD4 <250; i would either wait til CD4 was higher by ART, use one of the longer courses the CDC-recommendeds as in https://www.cdc.gov/tb/publications/ltbi/treatment.htm , or until other studies come out showing benefit.

As with HIV (and other communicable diseases), testing for and treating LTBI reflects the basic public health principle of treatment as prevention… 

geoff

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

For access to the dropbox, go to link: https://www.dropbox.com/sh/0bmvtita8mzms11/XDTwHySFFg

to get access to blogs since 8/15/17:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category

3. or you can just type in a name in the search box and get all the blogs with that name in them

to access older blogs from the BMJ website, from October 2013 until 8/15/17: go to http://blogs.bmj.com/bmjebmspotlight/category/archive/ 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list