| |
Safety and efficacy of switching to alternative nucleoside analogues following symptomatic hyperlactatemia and lactic acidosis
|
| |
| |
The purpose of this study published in the current issue of AIDS (AIDS 2003; 17(17):2495-2499) is to evaluate the safety and efficacy of rechallenging patients who have recovered from nucleoside reverse transcriptase inhibitor (NRTI)-induced symptomatic hyperlactatemia or lactic acidosis with alternative NRTI-containing regimens. The authors (J. Tyler Lonergan; R. Edward Barber; W. Christopher Mathews; Department of Medicine, University of California, San Diego Medical Center, California, USA.) found you could successfully rechallenge patients with different nukes.
Data in this case series was collected from patients followed at the UCSD Owen Clinic from July 1998 through September 2002. Cases of symptomatic hyperlactatemia were HIV-infected adults receiving NRTI who had symptoms compatible with hyperlactatemia and two lactates > 2 times the upper normal limit.
Lactic acidosis was defined as lactate > 5 mmol/l with bicarbonate < 20 mmol/l. The suspected offending NRTI in the prior regimen were replaced with other NRTI thought to have equivalent antiviral potency but less mitochondrial toxicity.
Ten patients diagnosed with symptomatic hyperlactatemia and two with lactic acidosis were later restarted on antiretrovirals that included new NRTI. The NRTI that patients were receiving when symptomatic hyperlactatemia or lactic acidosis was diagnosed included stavudine and lamivudine (n = 6), stavudine and didanosine (n = 4), and stavudine and abacavir (n = 2).
The median (range) peak lactate was 5.4 (4.7-19.1) mmol/l. Five patients were rechallenged with abacavir and lamivudine, five with zidovudine, abacavir and lamivudine, and two with zidovudine and lamivudine.
Among the 12 patients contributing over 22 years of cumulative reexposure to NRTI-containing therapy, one developed symptomatic hyperlactatemia again yielding a recurrence rate of 45.5 cases/1000 patient-years. Virologic control was maintained in all patients.
The authors concluded that this data supports the strategy that in cases of symptomatic hyperlactatemia or lactic acidosis in which the toxicity is associated with stavudine, didanosine or both, it is safe and efficacious to reintroduce NRTI that are less potent inhibitors of mitochondria.
Background
Hyperlactatemia is a frequent finding among HIV-infected individuals receiving nucleoside reverse transcriptase inhibitors (NRTI). In the majority of cases lactate elevations are transient and occur in the absence of symptoms. Symptomatic hyperlactatemia in contrast is an uncommon complication of NRTI, characterized by nonspecific, predominantly gastrointestinal symptoms, reproducible elevated lactate levels, and hepatic abnormalities which slowly resolve over weeks to months upon discontinuation of NRTI. Lactic acidosis represents the extreme form of NRTI-related hyperlactatemia syndromes and portends a poor prognosis. Although hyperlactatemia has been reported with all NRTI, it is most commonly associated with stavudine and didanosine.
The proposed mechanism of NRTI-induced hyperlactatemia is mitochondrial toxicity through inhibition of DNA polymerase gamma; the enzyme responsible for mitochondrial DNA synthesis. In vitro studies consistently demonstrate that NRTI differ in their ability to inhibit this enzyme and cause mitochondrial DNA depletion. The rank order of NRTI according to their ability to cause mitochondrial dysfunction in vitro is zalcitabine > didanosine > stavudine > zidovudine > abacavir = lamivudine = tenofovir.
Current treatment guidelines recommend that patients who have experienced symptomatic hyperlactatemia or lactic acidosis should subsequently be treated with NRTI-sparing antiretroviral regimens. For patients who have preexisting resistance to either protease inhibitors (PI) or non-nucleoside reverse transcriptase inhibitors (NNRTI) or both, it is difficult to maintain virologic control without the inclusion of NRTI. Since symptomatic hyperlactatemia and lactic acidosis are associated with mitochondrial dysfunction and some NRTI exhibit less mitochondrial toxicity than others in vitro, it might be safe to reintroduce NRTI that are more mitochondria toxicity-sparing. The objectives of this study were to evaluate the safety and efficacy of rechallenging patients with a prior episode of symptomatic hyperlactatemia or lactic acidosis with alternative NRTI-containing regimens.
Methods
The data in this case series were collected from chart review and a computerized database (Lab Tracker, Ground Zero Software, San Francisco, California, USA) of a large, urban, university-based HIV treatment center from July 1998 through September 2002. Symptomatic hyperlactatemia was defined as an HIV-infected adult receiving a NRTI-containing antiretroviral regimen with at least one of the following clinical manifestations: nausea/vomiting, abdominal pain, anorexia/weight loss, or fatigue plus two venous lactates elevated to > 2 times the upper normal limit (normal laboratory reference range for venous lactate is 0.7-2.1 mmol/l). A higher abnormal lactate threshold was used to define hyperlactatemia in this study to reduce the possibility of laboratory error and patient misdiagnosis. Lactic acidosis was defined as a lactate > 5 mmol/l with a bicarbonate < 20 mmol/l. Other etiologies for hyperlactatemia were excluded. Samples for lactate were collected at rest in fluoride-containing tubes, which were kept on ice until processed within 4 hrs h in the same laboratory.
When a patient was diagnosed with either symptomatic hyperlactatemia or lactic acidosis, all antiretroviral medications were discontinued. No other uniform intervention (e.g., riboflavin, antioxidants, dichloracetate) was implemented. On disappearance of symptoms associated with hyperlactatemia and improvement of lactate levels, patients were restarted on new NRTI-containing regimens. The NRTI suspected to have caused symptomatic hyperlactatemia or lactic acidosis were replaced with alternative NRTI thought to have equivalent antiviral efficacy but less propensity to injure mitochondria. For example, if a patient was taking a regimen that included stavudine and didanosine at the time they were identified with symptomatic hyperlactatemia or lactic acidosis, these two NRTI were switched to abacavir and lamivudine in the rechallenge regimen.
After resuming antiretroviral therapy, patients were monitored initially at 4 weeks and then every 12 weeks for a recurrence of lactic acidosis (as defined above) or symptomatic hyperlactatemia, which was defined more conservatively as the development of symptoms compatible with hyperlactatemia (as above) with any level of reproducible hyperlactatemia (lactate > 1 times upper normal limit). A recurrence rate of symptomatic hyperlactatemia was estimated as the number of cases of recurrent symptomatic hyperlactatemia per total person-time accumulated on the rechallenge regimens during the study period. Poisson confidence intervals (95% CI) were also computed.
Results
Twelve patients were diagnosed with either symptomatic hyperlactatemia or lactic acidosis and later rechallenged with modified NRTI-containing therapy . The median age of the group was 40 years (range, 28-51 years). Three-quarters of the patients were men. One half of them were Latino and the other half evenly divided between African-Americans and Caucasians. Two patients had chronic viral hepatitis (one had B and the other had C). Ten patients were on their first antiretroviral regimen when identified with symptomatic hyperlactatemia or lactic acidosis. The other two patients (7 and 8) had extensive prior exposure to other NRTI including zidovudine, zalcitabine, and lamivudine.
Symptomatic hyperlactatemia/lactic acidosis
All patients were receiving antiretroviral regimens containing two NRTI, which included stavudine at the time of diagnosis of symptomatic hyperlactatemia or lactic acidosis. Co-administered NRTI included lamivudine (n = 6), didanosine (n = 4) and abacavir (n = 2). Seven patients were also receiving a PI, one a NNRTI, and four both a PI and a NNRTI. The median time on these regimens when diagnosed was approximately 9.5 months. The median first and confirmatory lactate measurements were similar and both > 5 mmol/l. The second lactate was collected a median of 6 days (range 1-19 days) after the first measurement. Patient 1 who had the highest peak lactate level was hospitalized in the medical intensive care unit for fulminant lactic acidosis and while hospitalized developed ascending neuromuscular weakness. None of the other patients, including the other patient with lactic acidosis (7), required hospitalization. Alanine aminotransferase (ALT) levels were mildly to moderately elevated in nine patients (median ALT for entire group was 83 U/L; normal ALT, 10-45 U/L). Patients were generally well controlled on their regimens. Their median CD4 lymphocyte count was 326 cells/l and 10 patients had a HIV-1 RNA PCR < 400 copies/ml (Roche AMPLICOR PCR Standard 1.0 assay; Roche Diagnostics, Branchburg, New Jersey, USA).
NRTI withdrawal
Every patient stopped all antiretroviral agents after being identified with symptomatic hyperlactatemia or lactic acidosis. The median time off NRTI was 4 months. Patient 7 who had the longest period (> 1000 days) off NRTI before rechallenge was our first diagnosed case. This patient initially started an NRTI-sparing regimen, but he was unable to maintain virologic control on his therapy and was subsequently rechallenged with antiretrovirals that included NRTI. During the NRTI withdrawal period symptoms associated with hyperlactatemia resolved for all patients. Lactate levels improved for the entire group off therapy. The lactate was below 3 mmol/l for every patient prior to reintroducing antiretrovirals. ALT levels similarly improved in most patients and the median decreased to 48 U/L. As expected CD4 lymphocyte counts dropped to a median of 123 cells/l and the median viral load rebounded to over 180,000 copies/ml.
NRTI rechallenge
All patients were rechallenged with antiretroviral therapy that included either two or three NRTI. The only NRTI used were combinations of abacavir, zidovudine and lamivudine. Five were rechallenged with abacavir and lamivudine, five with abacavir, zidovudine and lamivudine, and two with zidovudine and lamivudine. Ten patients were also receiving either a PI (n = 5) or an NNRTI (n = 5) and two were receiving both. The median time on the rechallenge regimen was approximately 2 years. Two patients including patient 1 who had life-threatening lactic acidosis have been back on NRTI-containing therapy for 3 years. The median lactate among all rechallenged patients was normal. Four patients had elevated lactate levels at their last encounter. The highest lactate among this group was 3.3 mmol/l. Ten patients had normal ALT levels compared to only three at the time of symptomatic hyperlactatemia or lactic acidosis. The median ALT for the whole group was normal (25 U/L). CD4 lymphocyte counts increased to a median of 514 cells/l and the highest viral load in the group was 440 copies/ml. Only patient 9 had symptoms associated with hyperlactatemia. The remaining 11 patients including the other three patients with mild hyperlactatemia are asymptomatic.
Patient 9 who had the only recurrence was later rechallenged again with antiretrovirals. This time the patient restarted a regimen containing only two NRTI (abacavir and lamivudine). After 9 months back on therapy, she is asymptomatic, her lactate is normal and her viral load is undetectable (< 50 copies/ml).
Among the 12 rechallenged patients contributing 8110 days of cumulative reexposure to alternative dual and triple NRTI-containing therapy, one developed recurrent symptomatic hyperlactatemia and none developed lactic acidosis. The recurrence rate of symptomatic hyperlactatemia was estimated to be 45.5 cases per 1000 patient-years of NRTI reexposure (95% CI, 1.2-253.3 cases/1000 patient-years).
Discussion
Eleven of 12 patients with a prior episode of either symptomatic hyperlactatemia or lactic acidosis were successfully rechallenged with revised NRTI-containing antiretroviral regimens for a median of 2 years. NRTI (stavudine, didanosine) believed to have induced symptomatic hyperlactatemia or lactic acidosis were switched to alternative NRTI (zidovudine, abacavir, lamivudine) that have less propensity to cause mitochondrial dysfunction. After accumulating over 22 person-years of reexposure to NRTI, eight patients had normal lactate levels and no symptoms of hyperlactatemia, three had mild asymptomatic hyperlactatemia and one had a recurrence of milder symptomatic hyperlactatemia. No one developed lactic acidosis. Of equal importance all patients were able to regain virologic control of similar magnitude or better after resuming antiretroviral therapy.
The estimated recurrence rate of symptomatic hyperlactatemia was 45.5 cases per 1000 patient-years of NRTI rechallenge. Recurrence rates may be lower for patients rechallenged with two compared to three mitochondria toxicity-sparing NRTI. The only recurrent case in our series occurred in a patient who restarted three NRTI. We did not observe any recurrences for patients rechallenged with two NRTI.
Findings in this study are subject to three limitations: (i) precision of the recurrence rate estimates; (ii) potential patient misclassification, and (iii) selection bias. The small sample size limits the precision of our estimated recurrence rate of symptomatic hyperlactatemia, which is demonstrated by the wide 95% CI. Our data though is the most comprehensive to date. Other published reports of rechallenging patients who have a history of symptomatic hyperlactatemia or lactic acidosis with new NRTI include fewer patients (four or less) and shorter follow-up time (less than 6 months). Although these patients have also tolerated a strategy similar to ours the follow-up period in these case series may be too brief to observe a recurrence. The median follow-up time in this study of 12 patients was 2 years, which was greater than twice the median duration on the prior regimen that resulted in symptomatic hyperlactatemia or lactic acidosis.
Another limitation of this study is that some patients may have been misdiagnosed with symptomatic hyperlactatemia because of false lactate elevations and nonspecific symptoms. We minimized this possibility by including only patients with substantial, reproducible lactate elevations. Our case definitions of symptomatic hyperlactatemia and lactic acidosis are similar to the other published definitions and the ones currently used for AIDS Clinical Trials Group studies. Although the symptoms of hyperlactatemia are vague they are often distinguishable from common antiretroviral side effects because they occur after several months of therapy when most patients have overcome the initial adverse reactions of treatment.
Lastly the favorable results in this study may be due to patient selection bias. There are a few patients in our clinic who developed symptomatic hyperlactatemia who have not been rechallenged with NRTI. Because of sustained high CD4 lymphocyte counts off therapy these patients with their medical providers have chosen not to restart any antiretroviral medications. We also have had fatal cases of lactic acidosis. It is possible that these patients may not have tolerated a similar NRTI rechallenge strategy.
Patients who experience symptomatic hyperlactatemia or lactic acidosis face a difficult situation in the choice of subsequent antiretroviral regimens. Although the use of NRTI-sparing regimens remains the safest treatment strategy for these patients, this study supports that in cases in which this toxicity is associated with stavudine and didanosine, it is safe and efficacious to reintroduce NRTI (lamivudine, abacavir, and zidovudine) that are less potent inhibitors of mitochondria as another option. This latter approach is particularly valid for patients with limited treatment options for whom the inclusion of a NRTI backbone appears essential for the construction of a potent antiretroviral regimen. Other promising future NRTI rechallenge strategies include the use of tenofovir that appears safe in both in vitro mitochondrial toxicity studies and clinical trials and the use of adjunctive therapies (e.g., antioxidants, vitamin B supplements, dichloroacetate).
|
|
| |
| |
|
|
|
