Five years’ real-life experience with raltegravir in a large HIV centre

Setting

The Monsall Unit has around 2000 HIV-positive individuals regularly attending for care and has been a centre for HIV treatment since the first detection of HIV in the United Kingdom in the early 1980s. In general, BHIVA guidelines are followed and physicians in the department include co-authors of these guidelines.

Study methods

This was a retrospective, observational study evaluating routine clinical care. Pharmacy data and the departmental HIV database were used to identify all HIV-1 infected adults who had taken one or more doses of raltegravir, outside of clinical trials, from its availability (including expanded access) to 30 November 2012 at the Monsall Unit. We excluded those taking raltegravir as part of clinical trials as their data would already be published in those studies and because their management would reflect the clinical trial setting rather than routine clinical care. A standardised reporting form was used to collect demographic, clinical and laboratory data from electronic records, supplemented by paper records where necessary. The primary source of information was the detailed summary letter to the primary care physician (General Practitioner or GP) from each consultation. For a minority of individuals not registered with a GP, or not permitting correspondence with the GP, our practice is that letters are written in the same way but kept in hospital records only. Information recorded in routine clinical records was interpreted by investigators collecting data as far as was considered reasonable and ‘unknown’ recorded if data were not available or reasonable interpretation from clinical records not possible. Multiple reasons for starting or stopping raltegravir could be recorded per individual.

Duration on raltegravir was calculated from the start date to the time of the last contact with the patient when they were known to be taking raltegravir, usually the most recent outpatient visit. Mutations were analysed in the regional referral laboratory and sequences submitted to the Stanford database (http://hivdb.stanford.edu) for interpretation.

Study population

A total of 227 HIV-1 infected individuals were identified; 12 were excluded in total because insufficient clinical data were available (n = 7); raltegravir was used in a clinical trial (n = 2); identification details did not correspond to an existing patient (n = 2) and <18 years old (n = 1). Therefore, 215 individuals were included, with a median duration of 2.6 years on raltegravir use (interquartile range [IQR] 0.8, 3.5 years), giving 502 patient-years total raltegravir use. Raltegravir was stopped in 68 individuals, with the remaining 147 still taking raltegravir at the time of last contact.

Of these 215 individuals, 166 (77.2%) were male; median age was 45 years (IQR 38, 51) for males and 38 years (IQR 33, 43) for females; 144/166 (86.7%) of males and 10/48 (20.8%) of females were Caucasian and 19/166 (11.4%) of males and 35/48 (72.9%) of females were Black African; 189/215 (87.9%) were ART-experienced at the time of starting raltegravir (although 25 of these 189 were not taking ART at that time), with a median CD4 count of 323 cells/µL (IQR 132, 569) and 26/215 (12.1%) were ART-naive, with a median CD4 count of 54 cells/µL (IQR 23, 258).

Reasons for starting raltegravir

Among 26 ART-naive individuals, raltegravir was started for ‘rapid HIV viral load reduction’ in 14/26 (53.8%) to avoid drug–drug interactions (DDIs; 13/26 [50.0%]) or for tolerability reasons, including contraindications to efavirenz. Raltegravir was started during hospitalisation in 18/26, among whom the median CD4 count was 25 cells/µL (IQR 10, 95) and VL was 117,600 copies/mL (IQR 25,380, 642,081) compared with a median CD4 count of 330 cells/µL (IQR 226, 368) and VL of 18,033 (IQR 7100, 28,965) among those starting as an outpatient. Among 189 ART-experienced individuals, 217 reasons for selecting raltegravir were recorded, including 113/189 (59.8%) for tolerability reasons and 48/189 for previous virological failure (Figure 1). OPEN IN VIEWER

Accompanying drugs

In ART-naive individuals, raltegravir was co-administered with a fixed dose of tenofovir/emtricitabine in 17/26 and abacavir/lamivudine in 3/26. The remaining six took raltegravir in protease inhibitor (PI)-containing regimens, due to complex comorbidities (for rapid VL reduction; n = 5; 2/5 with suspected primary drug resistance) or contraindications to nucleos(t)ide reverse transcriptase inhibitors (NRTIs; n = 1). All six took ritonavir-boosted darunavir at the once-daily 800 mg darunavir/100 mg ritonavir dose as part of those PI-based regimens, and of these, four also had NRTIs as part of their combination therapy (one because of hepatitis B co-infection). Four of six on PI-based regimens started ART as inpatients with advanced HIV disease (baseline CD4 counts 45, 110 and 225 cells/µL [one baseline CD4 count unknown]). Of these four, three later stopped raltegravir to simplify their ART regimens because resistance test results were available (2/3) or because the VL became undetectable after discharge and the remaining regimen was considered sufficient.

Among ART-experienced individuals, raltegravir was co-administered with a dual NRTI backbone only in 53/189 (28.0%, with 31/40 taking fixed-dose tenofovir/emtricitabine; 9/40 fixed dose abacavir/lamivudine and 13/40 taking other NRTI combinations). Raltegravir was part of a PI-containing regimen in 116/189 individuals. The remaining 20 individuals had contraindications or resistance to PIs or NRTIs and took other combinations. Further detail of accompanying drugs is given in Table 1.

OPEN IN VIEWER

Table 1.

Accompanying regimens in antiretroviral therapy (ART)-naive and ART-experienced individuals starting raltegravir.

Regimen type	Detail	Number of individuals
ART-naive (N = 26)
NRTI backbone  only	Total	20
	TDF + FTC	17
	ABC + 3TC	3
PI-based	Total	6
	DRV/R 800/100 mg once daily	6/6
	(DRV/R plus NRTIs	4/6)
ART experienced (N = 189)
NRTI backbone  only	Total	53
	TDF + FTC	31
	ABC + 3TC	9
	Other	13
PI-based	Total	116
	DRV/R 800/100 mg once daily	20/116
	DRV/R 600/100 mg twice daily	37/116
	DRV/R other dose or unknown	1/116
	ATZ/R 300/100 mg once daily	38/116
	Other PI a	20/116
Among PI-based regimens, other accompanying drugs
	EFV (with ATZ/R + TDF) b	1/116
	NVP c	2/116
	ETV	31/116
	MRV d	6/116
	ENF e	4/116
Other regimens f	Total	20

Prior ART use

Data on previous ART exposure were available for 186/189 ART-experienced individuals. Previous PI-use was documented for 258/186 (84.9%) of ART-experienced individuals and previous enfuvirtide use in 21/186 (11.3%). Further detail of previous ART use is given in Table 2.

OPEN IN VIEWER

Table 2.

Previous ART use among 186 ART-experienced individuals starting raltegravir.

Drug class	Number (%) with previous exposure (denominator given where data not available for all individuals)
NRTI	186 (100)
First generation nNRTI	154/185 (83.2)
Second generation nNRTI	19/185 (10.2)
Protease inhibitors	158 (85.0)
Fusion inhibitors (enfuvirtide)	21 (11.3)

Reasons for stopping raltegravir

During this time period, 68 individuals stopped raltegravir (Figure 2). Among 18 individuals stopping raltegravir for toxicity reasons, one had acute hepatitis considered to be due to raltegravir, with alanine transaminase (ALT) levels increasing from 22 U/L at baseline to 540 U/L on day 30 of treatment. This Caucasian male was hepatitis B- and C-negative and raltegravir 400 mg twice daily was co-administered with tenofovir and ritonavir-boosted atazanavir, after resistance mutations were detected during a blip on tenofovir, abacavir and ritonavir-boosted atazanavir. Three individuals had headaches attributed to raltegravir and two stopped due to potential drug reactions (one, admitted with rash, angioedema and raised transaminases [ALT 314 U/L], with Epstein-Barr viraemia was taking darunavir/ritonavir concurrently with raltegravir and all drugs were stopped; the second was known to have multiple drug allergies and stopped raltegravir due to a rash, while also taking enfuvirtide and atazanavir/ritonavir). OPEN IN VIEWER

Twelve individuals stopped raltegravir for simplification, of whom four had taken raltegravir to avoid DDIs with antimycobacterial or antifungal treatment and stopped once those treatments were completed; two stopped on receipt of a resistance test permitting simplification; two had undetectable VL on PI-based regimens for several months and the stop was physician initiated and the remaining four requested simplification. Four individuals died (Figure 2). Among three inpatients, deaths were due to Pneumocystis pneumonia, tuberculous meningitis and toxoplasmosis; none considered related to ART. The fourth died at home from cardiac causes unrelated to his therapy.

Hepatitis co-infection

Thirty-five individuals with hepatitis B or C infection provided 92 patient-years of raltegravir use. Of these, 21 were positive for hepatitis B surface antigen (HBsAg) and 14 positive for hepatitis C RNA. The median ALT at baseline was 29 U/L, with range of 8–179 U/L. None stopped raltegravir due to hepatotoxicity.

Of 21 HBsAg-positive individuals, 19 were taking tenofovir-containing regimens and two took entecavir along with suppressive ART as tenofovir was contraindicated due to renal disease. Of 14 individuals with hepatitis C infection, three took pegylated interferon-alpha and ribavirin for hepatitis C while on raltegravir. None in this time period took directly-acting antivirals for hepatitis C.

Mycobacterial co-infection

Twenty-two individuals used raltegravir during treatment for mycobacterial co-infection, of whom 13/22 (59.1%) were female and 17/22 (77.3%) were Black African. The median CD4 count at the time of starting raltegravir was 83 cells/mm³ (IQR 23, 229) and, among 11/22 not on ART, median VL was 109,833 copies/mL (range 7766, 8,647,191). Three had known K103N mutation and one had G190S at baseline. Fourteen individuals were inpatients at the time of starting raltegravir and eight were ART naive.

The majority had M. tuberculosis (18/22 [81.8%]), including three with tuberculous meningitis and five with disseminated disease. Four (18.1%) had M. avium complex bacteraemia. Of the 22 individuals, 19 were taking rifampicin and two were taking rifabutin concurrently with raltegravir, while one was taking other antimycobacterial drugs for M. avium complex. Raltegravir was dosed at 800 mg twice daily in 13/18 with M. tuberculosis and 400 mg twice daily in five, and two of them commenced 800 mg twice daily then reduced after therapeutic drug monitoring. Only one stopped raltegravir due to side effects (peripheral neuropathy while also taking isoniazid) and two, one of whom had virological failure, switched therapy due to concerns about adherence and barrier to resistance. One individual with tuberculous meningitis died six weeks after starting raltegravir.

Pregnancy

Of six women taking raltegravir during pregnancy, one switched from a PI-based regimen for tolerability reasons affecting adherence and five started raltegravir to intensify PI-based regimens at median 32 weeks gestation (range 21–37 weeks, with the woman starting at 21 weeks previously taking ritonavir-boosted lopinavir only due to tolerability and extensive reverse transcriptase mutations). All six women had undetectable HIV RNA at delivery and all infants were HIV-negative with no foetal abnormalities.

Once-daily dosing

The off-license 800 mg once-daily dose was used in ten ART-experienced individuals. All were PI-experienced and five switched directly from PI-based regimens. Six individuals had non-nucleos(t)ide reverse transcriptase inhibitor resistance mutations (K103N, Y181C or both). Five individuals were off-treatment at the time of starting raltegravir and five were prescribed ART at that time, although only three had undetectable VL. Seven took once-daily raltegravir with a once-daily PI-based regimen (one without NRTIs) and the remaining three with NRTI backbone only. Details of regimens are given in Table 3.

OPEN IN VIEWER

Table 3.

Once-daily (OD) raltegravir (800 mg) in ten ART-experienced individuals.

Individual number	Regimen at time of starting RAL	Reason for starting RAL	Viral load at time of starting RAL (HIV RNA copies/mL)	CD4 count at time of starting RAL (cells/µL)	Documented adherence problems on RAL	RAL-containing regimen	Number of blips a on once-daily RAL	Outcome
1	Off treatment (previous PI use)	Tolerability	143,604	142	No	ABC + 3TC + RAL	N/A	Stopped all ART against medical advice
2	Off treatment (previous PI and ENF use)	Tolerability, resistance to other classes	134,921	85	Yes	TDF + ATZ/R + RAL	N/A	Frequent interruptions to therapy – did not suppress
3	Off treatment (previous PI use)	Tolerability, resistance to other classes	43,346	9	Yes	TDF + FTC + RAL	N/A	Stopped all ART against medical advice
4	Off treatment (previous PI use)	Resistance to other classes	66,141	137	Yes	TDF + ATZ/R + RAL	N/A	Stopped all ART against medical advice
5	TDF + ABC + ATZ/R	Tolerability, bone density	<40	368	No	ABC & ATZ/R + RAL	1	Virological suppression ongoing at 29.1 months
6	ZDV + 3TC + TDF + DRV/R	Tolerability, lipodystrophy, hyperbilirubinaemia	<40	184	Yes	TDF + DRV/R (800/100 mg od) + RAL	1	Virological suppression ongoing at 28.4 months
7	TDF + EFV + DRV/R	Tolerability, EFV side effects	<40	606	No	TDF + DRV/R (800/100 mg od) + RAL	0	Virological suppression ongoing at 27.3 months
8	TDF + ABC + LOP/R	Tolerability, high cardiovascular risk	59	1226	No	TDF + ABC + RAL	0	Virological suppression ongoing at 29.2 months
9	Off treatment (previous PI use)	Resistance to other classes	2720	17	Yes	TDF + FTC + DRV/R (800/100 mg od) + RAL	N/A	Stopped all ART against medical advice
10	TDF + FTC + DRV/R	Unknown	104	352	No	DRV/R (800/100 mg od) + RAL	0	Virological suppression ongoing at 7.2 months

On once-daily raltegravir, 4/10 stopped all ART against medical advice; one stopped due to virological failure with poor adherence after 26 months; one stopped raltegravir for simplification after 27 months and four remained suppressed on raltegravir-containing once-daily regimens after a median of 28 months.

Virological failure

Of 215 individuals, ten (4.7%) stopped raltegravir because of virological failure: of these, the HIV-1 integrase region was successfully sequenced in four and the remaining six had low-level viraemia precluding sequencing or failed sequencing. Two of the four individuals showed evidence of INSTI resistance after virological failure. The first had Q148R, on a background of pre-existing triple-class resistance; the second had Y143R, with pre-existing M184V, Y115F and major protease mutations. The two individuals without evidence of INSTI resistance had no pre-existing mutations in the reverse transcriptase or protease regions. Concerns about adherence were documented in 4/10.