|Home | About | Journals | Submit | Contact Us | Français|
Gram-positive bacterial infections are an important cause of morbidity and death among cancer patients, despite current therapy. In this case-control study, we evaluated the clinical outcomes and safety of telavancin in cancer patients with uncomplicated Gram-positive bloodstream infections (BSIs). Between March 2011 and May 2013, we enrolled cancer patients with uncomplicated Gram-positive BSIs to receive intravenous telavancin therapy for at least 14 days for Staphylococcus aureus and 7 days for other Gram-positive cocci. Patients with baseline creatinine clearance (CLCR) values of >50 ml/min received 10 mg/kg/day of telavancin, and those with CLCR values between 30 and 49 ml/min received 7.5 mg/kg/day. Patients were compared with a retrospective cohort of 39 historical patients with Gram-positive BSIs, matched for underlying malignancy, infecting organism, and neutropenia status, who had been treated with vancomycin. A total of 78 patients were analyzed, with 39 in each group. The most common pathogen causing BSIs was S. aureus (51%), followed by alpha-hemolytic streptococci (23%), Enterococcus spp. (15%), coagulase-negative staphylococci (8%), and beta-hemolytic streptococci (3%). Sixty-two percent of patients had hematological malignancies, and 38% had solid tumors; 51% of the patients were neutropenic. The overall response rate determined by clinical outcome and microbiological eradication at 72 h following the initiation of therapy, in the absence of relapse, deep-seated infections, and/or infection-related death, was better with telavancin than with vancomycin (86% versus 61%; P = 0.013). Rates of drug-related adverse events were similar in the two groups (telavancin, 31%; vancomycin, 23%; P = 0.79), with similar rates of renal adverse events. Telavancin may provide a useful alternative to standard vancomycin therapy for Gram-positive BSIs in cancer patients. (This study has been registered at ClinicalTrials.gov under registration no. NCT01321879.)
Gram-positive bacteria are among the most common pathogens responsible for health care-associated and nosocomial bacteremia (1). Vancomycin is commonly used to treat drug-resistant Gram-positive coccal bacteremia. However, the dosing of vancomycin may require adjustments according to the results of serum trough concentration monitoring. Furthermore, the performance of vancomycin may be reduced against methicillin-resistant Staphylococcus aureus (MRSA) strains that have vancomycin MICs of >1 mg/liter, although there is no convincing evidence that this effect translates into worse outcomes among patients (2,–4). In addition, vancomycin is inferior to beta-lactams in the treatment of methicillin-susceptible Staphylococcus aureus (MSSA) strains.
Telavancin is a semisynthetic lipoglycopeptide antimicrobial agent with bactericidal activity against susceptible Gram-positive pathogens, including MRSA isolates with vancomycin MICs of ≥2 μg/ml (5, 6). Telavancin has a dual mechanism of action; telavancin both inhibits cell wall synthesis and, unlike vancomycin, disrupts bacterial cell membrane function (7, 8). Telavancin is approved in the United States and Canada for treatment of complicated skin and skin structure infections caused by susceptible Gram-positive organisms, including methicillin-susceptible and methicillin-resistant S. aureus, vancomycin-susceptible Enterococcus faecalis, and other streptococci (9). It is also approved in Europe and the United States for treatment of hospital-acquired bacterial pneumonia (HABP) and ventilator-associated bacterial pneumonia (VABP) caused by susceptible isolates of S. aureus, when alternative treatments are not suitable (10). Telavancin is administered intravenously once daily (11, 12) and does not require dose adjustment based on the results of serum concentration monitoring.
There are limited data demonstrating the activity of telavancin in Gram-positive bacteremia. Therefore, the purpose of our study was to evaluate the feasibility of using telavancin for the treatment of Gram-positive bacteremia in cancer patients, including those with neutropenia.
(Portions of this study were presented in abstract form at IDWeek 2014, Philadelphia, PA, 8 to 12 October 2014.)
We conducted an open-label, single-institution pilot study (ClinicalTrials registration no. NCT01321879). The study was approved by the University of Texas M. D. Anderson Cancer Center institutional review board. Between March 2011 and May 2013, 40 consecutive cancer patients with uncomplicated Gram-positive bloodstream infections (BSIs) were prospectively enrolled in our study and received telavancin to treat Gram-positive BSIs. Eligible patients were ≥18 years of age and had uncomplicated Gram-positive bacteremia and at least 2 signs of sepsis within 48 h prior to the initiation of telavancin therapy. The following patients were ineligible for our study: patients who had received treatment with an antibiotic, such as vancomycin, linezolid, tigecycline, or telavancin, for more than 48 h within 72 h before the initiation of study medication administration; patients who had a deep-seated, intravascular source of infection with the same organism cultured from the blood (e.g., endocarditis or septic thrombosis); patients who had a prosthetic valve; patients who had an estimated serum creatinine clearance (CLCR) value of <30 ml/min; patients who had bilirubin levels higher than 4 times the normal upper limit (normal range, 0.2 to 1.3 mg/dl); and patients who had a history of hypersensitivity to lipoglycopeptides. One patient was excluded from the analysis because she had a positive serum β-human chorionic gonadotropin screening test result and did not receive telavancin.
Thirty-nine patients received at least 1 dose of telavancin and were included in the safety analysis. We compared the study patients with a retrospective cohort of 39 matched control cancer patients with uncomplicated Gram-positive BSIs who had been treated with vancomycin for at least 72 h. The control patients were matched with study patients according to underlying malignancy, type of Gram-positive infecting organism, and neutropenia status.
Telavancin was given intravenously for at least 14 days for S. aureus and 7 days for other Gram-positive cocci. Patients with baseline estimated CLCR values of >50 ml/min received a daily telavancin dose of 10 mg/kg, and those with CLCR values between 30 and 49 ml/min received a daily dose of 7.5 mg/kg. Patients with neutropenia who were at risk for Gram-negative sepsis and patients who developed polymicrobial bacteremia with Gram-negative organisms were treated with other broad-spectrum antimicrobial agents active against Gram-negative pathogens.
All patients were evaluated during treatment and at the end of treatment and were monitored for 1 month after the last dose of study drug. In the telavancin group, laboratory evaluations, including complete blood counts and blood chemistry tests, were performed every 7 days, and serum creatinine measurements were performed twice a week (on different days) for the first week of treatment and every 7 days thereafter until the end of therapy. Blood cultures were repeated every other day during treatment until negative results were obtained, at the end of treatment, and 1 month after the last dose of study drug. Patients were monitored for adverse events throughout the study. The control group consisted of historical patients with Gram-positive bloodstream infections who were treated with vancomycin and who were matched according to underlying disease, type of organism, and neutropenia status. The retrospective cohort did not follow the protocol schedule of events, and subjects were not closely monitored according to the protocol. In the control group, vancomycin dosing was not prespecified; however, the drug was administered to achieve goal trough levels of 15 to 20 μg/ml, in accordance with the published guidelines (17).
Our primary objective was to evaluate the safety and efficacy of telavancin. Additionally, patients treated with telavancin were compared to historical patients treated with vancomycin. For the efficacy endpoints, we evaluated clinical and microbiological responses, including resolution of signs and symptoms and microbiological eradication. Study patients were also monitored for the development of late complications, infection-related death, and/or relapse within the follow-up period. For the safety and tolerability endpoints, we evaluated the adverse events (including nephrotoxicity) that occurred during the study period for all patients who received at least one dose of the study antibiotic, and we determined the relationship of those adverse events to the study drug.
Uncomplicated bacteremia was defined as a positive blood culture result without evidence of persistent bacteremia or deep-seated infection (e.g., endocarditis, septic thrombosis, or osteomyelitis) present at enrollment. For positive blood cultures involving commensal skin organisms such as coagulase-negative Staphylococcus, Corynebacterium, Propionibacterium, Micrococcus, or Bacillus species, two positive blood cultures, or a single positive blood culture with either ≥100 colonies/ml or a time to positivity of ≤16 h (which has been shown to reflect high-grade bacteremia ), were required. The time to positivity of a blood culture was recorded in the microbiology laboratory by the automatic culture detector (Bactec 9240 system with Bactec Plus Aerobic/F medium; Becton Dickinson), which recorded culture positivity every 15 min according to changes in fluorescence related to microbial growth.
Clinical response was defined as the resolution of clinical signs and symptoms (mainly resolution of fever) within 72 h after the initiation of either telavancin or vancomycin therapy. Microbiological resolution was defined as eradication of the microorganism from the bloodstream within 72 h after the initiation of either telavancin or vancomycin therapy. Patients who did not have a repeat blood culture performed within the first 2 weeks after the initiation of appropriate study antibiotic therapy were excluded from this analysis. For the vancomycin arm, patients who experienced clinical resolution within 72 h but did not have repeat blood cultures performed within that time frame were considered to have microbiological resolution within 72 h, particularly if repeat blood cultures performed at a later time were negative. Persistence was defined as continued isolation of the Gram-positive organism from the blood 72 h after the initiation of appropriate antibiotic therapy. Relapse was defined as the recurrence of bacteremia during the follow-up period. An infection-related complication was defined as the development of a deep-seated infection that was not present or suspected at the onset of bacteremia but was subsequently diagnosed 1 week after the initiation of study antibiotic treatment. Overall response was defined as clinical and microbiological resolution within 72 h after the initiation of antibiotic therapy (telavancin or vancomycin), without evidence of relapse, infection-related complications, or infection-related death.
Nephrotoxicity was defined as a renal adverse event and was documented as a ≥50% increase in the serum creatinine level from the baseline level at any time during therapy. Neutropenia was defined as an absolute neutrophil count (ANC) of <500 cells/mm3.
All staphylococcal and enterococcal isolates were tested for susceptibility to vancomycin by using an automated Vitek 2 system (bioMérieux, Durham, NC). All other Gram-positive organisms were identified with the Etest method. All organisms from the patients in the telavancin group were tested for susceptibility to telavancin with the Etest method.
We used the chi-square test or Fisher's exact test to compare categorical variables, as appropriate. Continuous variables were compared using the Wilcoxon rank sum test, because of deviation of the data from a normal distribution. In addition, bivariate analyses were performed using the Cochran-Mantel-Haenszel test to assess the associations between treatment and some outcomes, with adjustment for potential confounders. All tests were two-sided tests, with a significance level of 0.05. Statistical analyses were performed using SAS version 9.3 (SAS Institute Inc., Cary, NC).
Thirty-nine cancer patients who were prospectively enrolled in our study and who received at least one dose of telavancin were included in the analysis. We compared the study patients with a retrospective cohort of 39 cancer patients with BSIs who were treated with vancomycin (Table 1). In both groups, the most common pathogen causing BSIs was S. aureus (51%), followed by alpha-hemolytic streptococci (23%), enterococci (15%), coagulase-negative staphylococci (8%), and beta-hemolytic streptococci (3%) (Table 2). Sixty-two percent of patients had hematological malignancies, and 38% had solid tumors. At the onset of bacteremia, 51% of the patients were neutropenic. The median ages and gender distributions were similar for the two groups (Table 1). At baseline, the groups had similar overall comorbidity rates; however, the patients in the telavancin group were more likely to have a history of acute renal insufficiency within 6 months before telavancin therapy (15% versus 0%; P = 0.025). In both groups, 15% of patients had undergone hematopoietic stem cell transplantation within 1 year before the onset of bacteremia. The rates of central-line-associated BSIs (CLABSIs) were similar in the two groups. The rates of central-line removal among patients with CLABSIs were also similar for the groups (Table 1). However, central venous catheters (CVCs) in CLABSIs were removed earlier in the telavancin arm than in vancomycin arm (a median of 2 days or 3.5 days, respectively, after the onset of bacteremia; P = 0.03) (Table 1).
The rates of polymicrobial bacteremia were similar in the two groups (10% in the telavancin group versus 8% in the vancomycin group; P > 0.99), with the following organisms being isolated: coagulase-negative staphylococci, enterococci, acid-fast bacilli, and Escherichia coli in the telavancin arm and Bacillus species and coagulase-negative staphylococci in the vancomycin arm. The rates of other sites of infection were also similar in the two groups (Table 1).
Patients treated with telavancin tended to have better clinical responses, including defervescence and an absence of deep-seated infections and relapse, than did patients treated with vancomycin (89% versus 72%; P = 0.09). Microbiological response rates were similar in the two groups (94.4% versus 92.1%; P > 0.99). Relapse, deep-seated infection, and overall mortality rates were comparable in the two groups. There were no infection-related deaths or drug-related deaths in either group. However, the overall response rate in the telavancin group was better than that in the vancomycin group (86% versus 61%; P = 0.013) (Table 3).
Based on the results of the univariate analyses described above, we performed further analyses of two outcomes, namely, clinical response and overall response. First, by univariate analysis we found that the following factors each had a potentially significant association with clinical response (P < 0.15): hematological malignancy, neutropenia, infection with S. aureus, and stem cell transplant within 1 year before the onset of bacteremia. Then we performed a series of bivariate analyses using the Cochran-Mantel-Haenszel test and found that none of those factors confounded the association between treatment and clinical response, which is consistent with our data showing that all of those factors were well balanced between the two groups (Table 1). Similar analyses were performed for overall response. As for clinical response, Cochran-Mantel-Haenszel tests showed that, of all of the baseline characteristics we evaluated, no factor confounded the association between treatment and overall response. Therefore, the Cochran-Mantel-Haenszel tests confirmed the findings for clinical response and overall response from the univariate analyses.
The rates of drug-related adverse events, including nephrotoxicity, were similar in the two groups (telavancin, 31%; vancomycin, 23%; P = 0.79). Nephrotoxicity occurred in 5 patients (12.8%) in the telavancin group and 6 patients (15.4%) in the vancomycin group (P = 0.74). The serum creatinine levels doubled from the baseline values in 2 of the 5 patients with nephrotoxicity in the telavancin group; in the other 3 patients, the serum creatinine levels increased from the baseline values by ≥50% but less than 100%. The highest creatinine levels in patients with nephrotoxicity were observed a median of 5 days (range, 3 to 17 days) after the initiation of telavancin therapy. Of the 5 patients with nephrotoxicity, 1 patient required reduction of the telavancin dose; for the other 4 patients, initially the telavancin dose was adjusted and then telavancin administration was discontinued. Three patients recovered renal function after telavancin administration was discontinued, and the other 2 patients died during the follow-up period, owing to the progression of their underlying malignancies. Median creatinine levels and CLCR values at baseline, at the end of treatment, and at the last follow-up visits were comparable in the two groups (Table 4).
In the telavancin group, the other drug-related adverse events included diarrhea (5.1%), altered taste (5.1%), nausea and vomiting (2.6%), skin rash (2.6%), anorexia (2.6%), palpitations (2.6%), and confusion (2.6%) (Table 5). These adverse events resolved with the discontinuation of telavancin administration and appropriate supportive care. Telavancin was discontinued because of drug-related adverse events for a total of 7 patients (18%); 4 patients had nephrotoxicity, 1 patient had skin rash, 1 had diarrhea, and 1 had confusion and tachycardia.
Dose adjustment occurred significantly more frequently in the vancomycin group than in the telavancin group (62% versus 13%; P < 0.0001). The only reason for adjusting the dose in the telavancin group was an increase in the creatinine level; however, vancomycin had to be adjusted owing to increased creatinine levels as well as low trough concentration levels. Of the 36 patients in the vancomycin group who had vancomycin trough concentrations assayed at least once (92% of the vancomycin patients), only 5 patients (14%) achieved recommended therapeutic vancomycin trough levels (≥15 μg/ml). Nineteen (68%) of the 31 patients with low vancomycin trough concentrations had at least a second vancomycin trough level measurement, and only 6 patients (32%) achieved recommended therapeutic vancomycin trough concentrations. In the vancomycin group, achieving recommended therapeutic vancomycin trough levels (≥15 μg/ml) did not affect the overall response rate (data not shown). In the vancomycin group, the rates of nephrotoxicity were similar among patients who achieved recommended therapeutic vancomycin trough levels (≥15 μg/ml) and those who did not (20% versus 16%; P>.99).
More than one-half of our isolates had vancomycin MICs of >0.5 mg/liter. All of the telavancin isolates tested had telavancin MICs of <0.5 mg/liter (Table 2).
This is the first study to evaluate telavancin for the treatment of uncomplicated Gram-positive BSIs in cancer patients, including those with neutropenia. Patients treated with telavancin had clinical response and microbiological eradication rates similar to those of matched historical patients treated with vancomycin. Furthermore, patients treated with telavancin had better overall response (clinical response and microbiological eradication without relapse or infection-related complications) rates than did those treated with vancomycin (86% versus 61%; P = 0.013). This finding could be partly attributed to the dual mechanism of action of telavancin and its rapid bactericidal activity. Like vancomycin, telavancin inhibits cell wall synthesis of Gram-positive organisms. Unlike vancomycin, however, telavancin also disrupts the plasma membrane function of pathogens, which could explain the potent antibacterial activity of telavancin against Gram-positive bacteria (5,–8, 14).
Our results are similar to the results of a recent randomized trial in which patients with S. aureus bacteremia were randomly assigned to receive either telavancin or standard therapy (vancomycin or antistaphylococcal penicillin); that study showed a response rate of 88% in the telavancin group (15). The previous trial focused on S. aureus bacteremia, however, and had only 8 clinically evaluable patients in the telavancin arm, whereas our study's telavancin group included 36 clinically evaluable patients with cancer and with different Gram-positive organisms known to cause BSIs, including 20 patients with S. aureus.
We found that the rate of drug-related adverse events was 31%. The rate of drug-related adverse events that led to discontinuation of telavancin treatment was 18% in our study, which is higher than rates found previously in different randomized controlled trials comparing telavancin and vancomycin for the treatment of complicated skin and skin structure infections and hospital-acquired pneumonia, as well as the recent bacteremia study (9, 10).
We observed a nephrotoxicity rate of 12.8%, which is slightly higher than the rate reported previously by Rubinstein et al. for 751 patients with hospital-acquired pneumonia treated with telavancin and higher than the rate reported by Polyzos et al. in a meta-analysis of 6 randomized controlled trials comparing telavancin with other antibiotics (10, 16). The nephrotoxicity rate found in our study is also higher than that reported in the recent bacteremia study (15). In the other studies, nephrotoxicity was defined as an increase in the serum creatinine level of ≥1.5 mg/dl and a creatinine level at least 50% greater than the baseline level at any time during the treatment period. In our study, we might have been more vigilant, and we defined nephrotoxicity as a serum creatinine level that was at least 50% more than the baseline value (for example, a serum creatinine level of 1.2 mg/dl would have met our definition of nephrotoxicity if the patient's baseline serum creatinine level had been 0.8 mg/dl). This difference in the definitions of nephrotoxicity may account for the different results.
One of the advantages of telavancin is the convenient mode of administration (given once daily); vancomycin requires multiple daily doses. In addition, telavancin does not require monitoring of serum trough concentrations, whereas trough concentration measurements are essential for guiding vancomycin dosing, particularly for serious infections such as bacteremia (17). Therefore, telavancin may be more convenient for home health care-based therapy and also may improve patient compliance.
Although the Infectious Diseases Society of America (IDSA) clinical practice guidelines for treating MRSA infections in adults and children recommend vancomycin trough concentrations of 15 μg/ml for treating serious infections such as bacteremia (17), achieving the targeted concentrations may be challenging in clinical practice, as demonstrated by the findings for the retrospective cohort in our study; targeted levels were achieved for only 14% of the patients before the fourth vancomycin dose. This could have led to a better overall response rate in the telavancin group. The guidelines do not recommend the use of telavancin as a first-line treatment for MRSA bacteremia but do recommend telavancin for managing persistent MRSA bacteremia in cases with reduced susceptibility to vancomycin and daptomycin (17).
Our study showed that telavancin could be considered a convenient alternative to vancomycin for treating patients with uncomplicated bacteremia caused by Gram-positive cocci. However, close monitoring of renal function is recommended. As in previous studies, we found that telavancin inhibited all of our isolates, including those with vancomycin MICs of ≥1 μg/ml (6).
Our study had several limitations. First, the major limitation was the retrospective nature of our control group. Patients treated with telavancin were prospectively enrolled and closely monitored according to the protocol. However, the control group consisted of a retrospective cohort of cancer patients who had received vancomycin for treatment of their Gram-positive coccal bacteremia. Although the patients in the control group were matched with study patients according to underlying malignancy, bacteria, and neutropenia status, the control patients were not closely monitored according to the protocol and did not follow a specific schedule. Clinical signs and symptoms were not closely monitored, and serial blood cultures were not always available to document bacterial eradication. Second, the assessment and attribution to vancomycin of adverse events in the retrospective cohort may have been underestimated. Third, a small number of patients were included in our study. Despite the great challenge in bacteremia trials in enrolling patients with uncomplicated Gram-positive bacteremia, particularly cancer patients with neutropenia, our pilot study is the first and largest study to evaluate the role of telavancin in this patient population.
In conclusion, telavancin may offer a therapeutic alternative for treating bacteremia caused by susceptible Gram-positive bacterial isolates in cancer patients; in fact, telavancin treatment may result in improved overall response rates, compared with vancomycin, although the limitations of our control group should be considered. Telavancin appears to be a convenient alternative to vancomycin; our results showed that the telavancin group had a lower frequency of dose adjustment than did the vancomycin group but with a similar safety profile in this patient population. Telavancin should be used cautiously, and renal function should be monitored closely. A large randomized controlled trial including patients with complicated BSIs is warranted.
We acknowledge and dedicate our work to our deceased colleague Munirah Alshuaibi, for her contribution to the data collection.
Funding for the study was provided by both Theravance, Inc., and Astellas Pharma Global Development; telavancin was provided by Theravance, Inc.
We have no conflicts of interest to declare.
This research was supported in equal parts by funding from Theravance, Inc., and Astellas Pharma Global Development, Inc., to The University of Texas M. D. Anderson Cancer Center under project number CS2010-00032796.