ExcelMale
Menu
Home
What's new
Latest activity
Forums
New posts
Search forums
What's new
New posts
Latest activity
Videos
Lab Tests
Doctor Finder
Buy Books
About Us
Men’s Health Coaching
Log in
Register
What's new
Search
Search
Search titles only
By:
New posts
Search forums
Menu
Log in
Register
Navigation
Install the app
Install
More options
Contact us
Close Menu
Forums
Testosterone Replacement, Low T, HCG, & Beyond
Testosterone and Men's Health Articles
F.D.A. Panel Backs Limits on Testosterone Drugs but Rejects Petition for More Regulations
JavaScript is disabled. For a better experience, please enable JavaScript in your browser before proceeding.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an
alternative browser
.
Reply to thread
Message
<blockquote data-quote="Nelson Vergel" data-source="post: 14667" data-attributes="member: 3"><p>Vigen et al. conducted a retrospective cohort study to assess the association between testosterone therapy and all-cause mortality, MI, and stroke in men with low testosterone levels (<300 ng/dL) who had undergone coronary angiography between the years 2005 and 2011. The criteria for coronary artery disease as described in this publication was 2::20% lesion in any epicardial artery. Of 23,173 candidates assessed, a total of 8,709 men were evaluated (1,223 receiving testosterone and 7,486 not receiving testosterone). The average follow-up was about 840 days (540 days for the testosterone group versus 889 days for the no-testosterone group). Among the 14,464 men excluded, 128 were excluded because they sustained an MI or stroke prior to being prescribed testosterone.</p><p> </p><p>There were a total of 1,710 events (testosterone group: 67 deaths, 23 Mis, 33 strokes; no* testosterone group: 681 deaths, 420 Mis, 486 strokes). For each event, the percentage of occurrence was consistently lower in the testosterone group versus the no-testosterone group, respectively (death: 5.5% versus 9.1%; MI: 1.9% versus 5.6%; stroke: 2.7% versus 6.5%). The authors determined that the association between testosterone and MI, stroke, and death was significant (hazard ratio (HR) 1.29; 95% CI 1.04-1.58) when testosterone therapy was evaluated in an adjusted Cox proportional hazards model with stabilized inverse probability of treatment weighting and treating testosterone as a time varying covariate. The variables used to create the weights were a long list of demographics, concomitant disease, and procedures. After these adjustments, the incidences of death, stroke, and MI were numerically higher in the testosterone group versus the no-testosterone group, but were not statistically significant. Further, the effect of testosterone on cardiovascular risk was harmful in the adjusted analysis (incidence rates: 25.7% for testosterone vs. 19.9% for no-testosterone), but was beneficial in the unadjusted analysis (incidence rates: I 0.0% for testosterone vs. 21.1% for no-testosterone).</p><p> </p><p>The strengths of the Vigen study are the size of the database and the linkage to lab results. The study used the nationwide Veterans Affairs system and thus had a large sample size of middle* aged and older men (the mean age for the testosterone group was 61 and was 64 for the no* testosterone group). However, the study was the source of some controversy in the academic community and the authors have acknowledged some key weaknesses associated with this study, including unknown time of day in which blood levels were drawn for testosterone measurement and lack of endpoint validation due to the retrospective nature of the study. FDA also identified a number of limitations with the study.</p><p> </p><p>To start, the Vigen study did require patients to have a low serum testosterone to qualify for inclusion. However, the adequacy of testosterone treatment received is questionable. Only 60% of study patients had a follow-up serum testosterone level evaluated after starting testosterone therapy. On average, the subjects had a baseline testosterone level of 175.5 ng/dL and post treatment level of 332.2 ng/dL, which, while technically within the normal range, is below the target range generally used for treatment (400-500 ng/dL).32 Also, with an average of 332 ng/dL, a substantial number of patients in this study were likely to have had values below the lower limit of normal (300 ng/dL) and to have remained hypogonadal. This is understandable as 63% of study subjects used testosterone patches, a dosage form which often results in testosterone levels in the low normal range. The adequacy of testosterone treatment seen in this study does not appear to reflect the recommended clinical guidelines for testosterone replacement. Due to these treatment uncertainties, it is difficult to attribute the increased risk for the composite outcome to testosterone therapy alone. It is important to consider that the study subjects might have simply remained hypogonadal and thus at higher risk for cardiovascular events, regardless of treatment.</p><p> </p><p>The Vigen study also did not include any information on why some patients with low testosterone were treated and others were not treated. Overall, the study population had a high comorbidity burden; however, patients treated with testosterone tended to be younger, have lower testosterone levels, be more overweight, and, generally, have a lower comorbidity burden compared to the no-testosterone group. The criteria for coronary artery disease lacked clinical correlation and relied solely on a subjective angiographic measurement. These selection criteria likely created a cohort of widely variable clinical presentation which may not have been equitably distributed between the arms of the study. Although Vigen eta!. used weights to adjust for these imbalances, the two cohorts may not have been balanced with respect to underlying cardiac risk potential, and differences in cardiac outcomes may be due to this imbalance.</p><p> </p><p>In addition, the use of a composite outcome makes the interpretation of the study results less clear. Despite the data showing no significant difference between the testosterone and no* testosterone groups in the incidence of MI, stroke, and death, the authors' conclusions were contradictory to the crude event rates and were based on a complex statistical model using a weighting scheme involving approximately 50 variables. At least one key variable was not accounted for: the significantly lower baseline testosterone level in the testosterone group, which was viewed by the authors' peers as a significant oversight. 33</p><p> </p><p>Finally, the exclusion of 128 patients who experienced MI or stroke before initiating testosterone was not appropriate. These patients should have been included in the analysis and their events included in the no-testosterone group, which would have raised the event rate in the no* testosterone group by 71%. Their exclusion biased the results by reducing the number of events in the no-testosterone group. This issue was also raised by other researchers in their comments regarding this article.3 The authors issued a correction stating that, while their original publication noted that 1,132 patients had been excluded, this was "an incorrect notation" and only 128 patients were excluded for having an MI before testosterone therapy was initiated.35</p><p>Although the authors claimed that including the 128 patients did not change the result (HR 1.3; 95% CI 1.1-1.6), their exclusion raises significant concerns about the quality of the study. It is also unclear why the authors excluded 1,301 participants for not having coronary anatomy data (CAD status), considering the wealth of baseline information collected on medical and drug history. It is unclear how this exclusion might also have affected the risk estimate.</p><p> </p><p>Given the described limitations of the study by Vigen et al. it is difficult to attribute the reported findings to testosterone treatment.</p><p></p><p></p><p><u>Finkle et al</u></p><p> </p><p>Finally, the Petition relies on the outcome of a 2014 study by Finkle eta!. The Petition claims that "the statistically significant findings were that the risk of heart attacks after using the drug [testosterone] for three months was twice the risk in the year before use in all men 65 and over. Furthermore, the study found for the first time that in those men under 65 with a history of heart disease, there was 2.9-fold increase in heart attack risk."36</p><p> </p><p>Finkle conducted a retrospective cohort study to assess a possible association between testosterone therapy and non-fatal MI 90 days following an initial prescription. The Finkle study had the largest sample size of the four studies reviewed. Finkle used the MarketScan database to identify 55,593 patients prescribed testosterone and 167,279 patients prescribed a phosphodiesterase-S inhibitor (PDESI). The authors first used a self-control cohort method to compare the post-exposure incidence rate with the pre-exposure rate among the testosterone exposed cohort, and then used a parallel cohort method to compare the incidence rates between testosterone and PDESI patients. The results from the self-control analysis showed that for 55,593 patients initiating testosterone, there was a significant increased risk within the first 90 days following initiation of testosterone compared with the risk in one year preceding testosterone therapy (relative risk (RR) 1.36; 95% CI 1.03, 1.81). The subgroup analysis of age and history of heart disease in the testosterone patients also showed significant increased risks for post- versus pre-exposure for participants older than 65 years of age with no heart disease (RR 2.21; 95% CI 1.09-4.46), and less than 65 years of age with heart disease (RR 2.90; 95% CI 1.49-5.62). The results from the parallel cohort analysis were similar. Males in the testosterone cohort had an increased risk for non-fatal MI, ratio of the rate ratios (RRR) (1.27; 95% CI 0.94- 1.71). There was an approximate doubling of the risk for males older than 65-years old with heart disease (RRR 1.90; 95% CI 0.66-5.50) and without heart disease (RRR 2.41; 95% CI 1.12- 5.17), and for males younger than 65-years old with heart disease (RRR 2.07; 95% CI 1.05- 4.11).</p><p> </p><p>These data suggested a significant risk of MI in all patients prescribed testosterone, driven by those patients more than 65 years of age and those patients less than 65 years of age with a history of heart disease. For those patients older than 65 years of age with a history of heart disease taking testosterone, there was no significant risk for MI. Despite these results, the study has some limitations that raise questions about whether there is a true risk for non-fatal MI with testosterone therapy.</p><p> </p><p>The large size of the MarketScan administrative database allowed the investigators to evaluate the risk among a large number of patients. However, in these data, the diagnostic indications for testosterone were not available. Further, results of laboratory testing of testosterone levels are not available; this may be important as low serum testosterone is a known risk for cardiovascular events. Testosterone exposure was determined based on a patient's filling of a prescription for testosterone therapy, but it is unknown whether the patient actually used the prescription. This fact, combined with the inability to assess baseline or post-treatment testosterone levels or indication for therapy associated with the males treated with testosterone, makes it impossible to determine if the testosterone levels in these treated males had reached therapeutic range. Due to</p><p>these limitations, it is difficult to completely attribute the increased risk for non-fatal MI to testosterone treatment.</p><p> </p><p>It is also questionable as to whether the self-controlled cohort or the active comparator design is most appropriate for assessing outcomes of testosterone therapy. Normally, testosterone is prescribed for chronic use, but Finkle limited follow-up to 3-months of therapy. It is unclear if 3-months' follow-up is adequate to capture the relevant outcomes.</p><p> </p><p>In the testosterone cohort, the males tended to be younger with a higher comorbidity load compared with males in the PDE5I cohort. For both the crossover analyses and the active comparator analysis, the overall risk was small. However, age and heart disease status appear to be confounding factors. In both the crossover and active comparator analyses, the authors found a 2-fold increased risk in males over 65-years old without heart disease and a 2-3 fold increased risk in males with heart disease regardless of age.</p><p> </p><p>In addition, acute non-fatal MI was the only outcome measured. The study subject would have had to survive to be included in the analysis. As fatal MI and other outcomes such as cardiovascular mortality or stroke were not captured, it is unclear how their inclusion would have affected the study results.</p><p> </p><p>Due to these uncertainties, it is difficult to attribute the increased risk for non-fatal MI seen in the Finkle study to testosterone alone and not consider that the study participants might have remained hypogonadic and thus at higher risk for non-fatal MI.</p><p> </p><p><u>Other Literature</u></p><p> </p><p>In addition to reviewing the studies cited in the Petition and the literature search discussed above, FDA also performed a literature search to identify other articles that may be relevant to the question of whether testosterone can be linked to increased cardiovascular risk. We identified two relevant studies, which show either an apparent benefit of treatment with testosterone or an inference that testosterone therapy is not associated with an increased cardiovascular risk. The</p><p>first, a 2012 study by Shores et a!.,37 is an observational study designed to examine the</p><p>association between testosterone treatment and mortality in men with low testosterone. The database included seven Northwest Veterans Affairs medical centers and included a cohort of 1,031 male veterans older than 40 years of age with low testosterone (<250ng/dL). In this study, testosterone treatment was associated with a decreased mortality compared with no testosterone treatment (HR 0.61; 95% CI 0.42-0.88).</p><p> </p><p>Similarly, in a 2013 prospective follow-up from a previously reported cohort that collected data from outpatient medical facilities with access to medical records, Muraleedharan et a!.38 concluded that low testosterone levels predicted an increase in all-cause mortality during long* term follow-up, and testosterone replacement may improve survival in hypogonadal men with type 2 Diabetes Mellitus.</p><p> </p><p> </p><p>The Shores study used a lower threshold for low testosterone (<250 ng/dL) to increase the likelihood of a symptomatic hypogonadal population. In both studies, lower testosterone level was associated with testosterone therapy, but Body Mass Index and younger age were also predictors of use of testosterone in the Shores study. Otherwise, comorbid conditions were balanced between the cohorts in each of the studies. Both studies used time-to-event Cox regression analysis to calculate the risk estimates and confidence intervals, and they each showed an approximate 50% reduction in the risk for death with testosterone therapy.</p><p> </p><p>In addition, the Shores study showed that increasing mortality is associated with lower baseline testosterone levels and shorter duration of testosterone therapy. This might indicate that sicker, undertreated males are at a higher risk for mortality. The Muraleedharan study showed a similar mortality rate for the diabetic males on testosterone and for a cohort of diabetic males who had normal testosterone levels and no testosterone therapy, which suggests the untreated males in this study have similar outcomes to the treated hypogonadal males. On average, duration of testosterone therapy was longer than two years for the majority of the treated subjects, and the peak testosterone level was 657 ng/dL. Over 67% of the treated males had a testosterone level of over 518 ng/dL. These levels are well within the recommended mid-to-normal therapeutic levels (400 -700 ng/dL). In both studies, only all-cause mortality was reported as the outcome, so outcomes of interest that did not result in death were not captured.</p><p> </p><p><u>FDA's Response</u></p><p> </p><p>As discussed in this response, the studies presented in the Petition have significant limitations</p><p>that weaken their evidentiary value for confirming a causal relationship between testosterone and adverse cardiovascular outcomes. These weaknesses include:</p><p> </p><p>• Short follow-up times precluding assessment of the potential for long term benefits of testosterone therapy (Finkle);</p><p>• Unclear statistical methods (Vigen);</p><p>• Inability to compare results across studies due to differing outcomes and populations (Vigen, Finkle);</p><p>• Overall effect estimates are small and may be due, in part, to residual confounding (Vigen, Xu, Finkle);</p><p>• Limitations with respect to ascertainment of events (Basaria, Finkle);</p><p>• Overly broad case definition for cardiovascular events (Xu);</p><p>• Incomplete or unavailable laboratory data to confirm hypogonadism or to assess whether patients returned to normal testosterone levels after receiving treatment (Finkle, Vigen);</p><p>• Failure or inability to assess other potentially relevant laboratory data such as hematocrit or hemoglobin (Finkle);</p><p>• Non-validated endpoints or lack of compliance data (Finkle, Vigen, Xu); and</p><p>• Conflicted results suggesting both a testosterone benefit (Shores and Muraleedharan) and testosterone harm with respect to cardiovascular risk, or no difference between groups (Vigen, Xu).</p><p> </p><p>In addition, FDA has identified other studies in the literature that contradict the findings in the studies submitted.</p><p></p><p></p><p>Prior to the submission of the Petition, the Agency had already undertaken a thorough evaluation of the literature and other evidence to determine if additional regulatory action is necessary to protect consumers from the cardiovascular risks of testosterone therapy. As our January 31, 2014, drug safety communication indicated, FDA believes that the publication of these studies warrants further exploration of a possible safety signal regarding testosterone and cardiovascular risk. Our current evaluation remains ongoing. For the reasons discussed above, the Agency does not believe at this time that the evidence presented in the Petition is sufficient to require the addition of a boxed warning regarding cardiovascular risks of testosterone therapy to the labeling of all testosterone products. Therefore, at this time, FDA declines to exercise its authority to require safety labeling changes regarding these risks on the basis of the evidence presented in the Petition, and your request is denied. Consequently, your requests that FDA require FDA* approved Medication Guides for testosterone products to be updated with the same warnings and that manUfacturers be required to send Dear Doctor Letters regarding these risks are also denied.</p><p> </p><p>We are continuing to assess this potential safety signal. In particular, we are awaiting the results of the Testosterone Trial,39 a multicenter study of six coordinated trials investigating the effects of testosterone treatment in elderly men with low testosterone on physical function, vitality, sexual function, cognitive function, anemia, and cardiovascular risk. Eight hundred men over 65-years-of age whose serum testosterone is less than 250 ng/dL have been or will be randomized to receive testosterone or placebo double blindly for one year. Although this trial is not a safety study, we believe that the data will yield important information regarding the safety of testosterone with regard to cardiovascular risks. In addition, we intend to present the question of the potential association between testosterone and adverse cardiovascular events to an Advisory Committee this fall.</p><p> </p><p>Based on the outcome of these efforts, FDA intends to make a determination as to whether any regulatory action is warranted, such as invoking our authority to require safety labeling changes under section 505(o)(4) of the FD&C Act for testosterone-containing drugs, as appropriate.</p><p> </p><p><strong>III. CONCLUSION</strong></p><p> </p><p>After careful consideration, and, in light of the foregoing, we hereby deny your Petition in its entirety. FDA will continue to evaluate the cardiovascular risks of testosterone, and, if warranted, will take appropriate regulatory action to protect the public health when its evaluation has concluded.</p><p> </p><p>Janet Wood****, M.D. Director</p><p>Center for Drug Evaluation and Research</p><p> </p><p> </p><p> </p><p></p><p></p><p>29 Protocol: Testosterone and cardiovasclar related events in men: a meta.analysis of randomized controlled trials.</p><p>12-5-2011. <a href="http://www.crd.york.ac.uk/PROSPEROFILES/1815" target="_blank">http://www.crd.york.ac.uk/PROSPEROFILES/1815</a> <u>PROTOCOL 20111108.pdf.</u></p><p>30 Petition at 4.</p><p>31 Incorrect Number of Excluded Patients Reported in the Text and Figure. <em>Journal of American Medical Association </em>2014, 311: 967. (Incorrect Number); Comment and Response. <em>Journal of American Medical Association </em>2014, 311: 964-965. (Comment and Response).32 The Endocrine Society Clinical Practice guidelines recommend raising serum levels to between 400 and 700 ng/dL. See Bhasin S, Cunningham GR, Hayes FJ et al. Testosterone therapy in men with androgen deficiency syndromes: an Endocrine Society clinical practice guideline. <em>J </em><em>C/in Endocrinol Metab </em>2010;95:2536-2559.</p><p>33 Traish, A, eta!., 2014, Death by testosterone? We think not! <em>J </em><em>Sex Med, </em>11:624-629.</p><p>34 Comment and Response, 2014.</p><p>35 Incorrect Number, 2014.36 Petition at 5.37 Shores, M, et al., 2012, Testosterone treatment and mortality in men with low testosterone levels, <em>J </em><em>Clinical</em></p><p><em>Endocrinol Metab, </em>97 (6):2050-2058.</p><p>38 Muraleedharan, V, et al., 2013, Testosterone deficiency is associated with increased risk of mortality and testosterone replacement improves survival in men with type 2 diabetes, <em>Eur </em><em>J </em><em>Endocrino/, </em>169: 725-733.</p><p>39 For more information regarding the Testosterone Trial, see <a href="http://www.med.upenn.edu/idom/t-trial.html" target="_blank">http://www.med.upenn.edu/idom/t-trial.html.</a></p></blockquote><p></p>
[QUOTE="Nelson Vergel, post: 14667, member: 3"] Vigen et al. conducted a retrospective cohort study to assess the association between testosterone therapy and all-cause mortality, MI, and stroke in men with low testosterone levels (<300 ng/dL) who had undergone coronary angiography between the years 2005 and 2011. The criteria for coronary artery disease as described in this publication was 2::20% lesion in any epicardial artery. Of 23,173 candidates assessed, a total of 8,709 men were evaluated (1,223 receiving testosterone and 7,486 not receiving testosterone). The average follow-up was about 840 days (540 days for the testosterone group versus 889 days for the no-testosterone group). Among the 14,464 men excluded, 128 were excluded because they sustained an MI or stroke prior to being prescribed testosterone. There were a total of 1,710 events (testosterone group: 67 deaths, 23 Mis, 33 strokes; no* testosterone group: 681 deaths, 420 Mis, 486 strokes). For each event, the percentage of occurrence was consistently lower in the testosterone group versus the no-testosterone group, respectively (death: 5.5% versus 9.1%; MI: 1.9% versus 5.6%; stroke: 2.7% versus 6.5%). The authors determined that the association between testosterone and MI, stroke, and death was significant (hazard ratio (HR) 1.29; 95% CI 1.04-1.58) when testosterone therapy was evaluated in an adjusted Cox proportional hazards model with stabilized inverse probability of treatment weighting and treating testosterone as a time varying covariate. The variables used to create the weights were a long list of demographics, concomitant disease, and procedures. After these adjustments, the incidences of death, stroke, and MI were numerically higher in the testosterone group versus the no-testosterone group, but were not statistically significant. Further, the effect of testosterone on cardiovascular risk was harmful in the adjusted analysis (incidence rates: 25.7% for testosterone vs. 19.9% for no-testosterone), but was beneficial in the unadjusted analysis (incidence rates: I 0.0% for testosterone vs. 21.1% for no-testosterone). The strengths of the Vigen study are the size of the database and the linkage to lab results. The study used the nationwide Veterans Affairs system and thus had a large sample size of middle* aged and older men (the mean age for the testosterone group was 61 and was 64 for the no* testosterone group). However, the study was the source of some controversy in the academic community and the authors have acknowledged some key weaknesses associated with this study, including unknown time of day in which blood levels were drawn for testosterone measurement and lack of endpoint validation due to the retrospective nature of the study. FDA also identified a number of limitations with the study. To start, the Vigen study did require patients to have a low serum testosterone to qualify for inclusion. However, the adequacy of testosterone treatment received is questionable. Only 60% of study patients had a follow-up serum testosterone level evaluated after starting testosterone therapy. On average, the subjects had a baseline testosterone level of 175.5 ng/dL and post treatment level of 332.2 ng/dL, which, while technically within the normal range, is below the target range generally used for treatment (400-500 ng/dL).32 Also, with an average of 332 ng/dL, a substantial number of patients in this study were likely to have had values below the lower limit of normal (300 ng/dL) and to have remained hypogonadal. This is understandable as 63% of study subjects used testosterone patches, a dosage form which often results in testosterone levels in the low normal range. The adequacy of testosterone treatment seen in this study does not appear to reflect the recommended clinical guidelines for testosterone replacement. Due to these treatment uncertainties, it is difficult to attribute the increased risk for the composite outcome to testosterone therapy alone. It is important to consider that the study subjects might have simply remained hypogonadal and thus at higher risk for cardiovascular events, regardless of treatment. The Vigen study also did not include any information on why some patients with low testosterone were treated and others were not treated. Overall, the study population had a high comorbidity burden; however, patients treated with testosterone tended to be younger, have lower testosterone levels, be more overweight, and, generally, have a lower comorbidity burden compared to the no-testosterone group. The criteria for coronary artery disease lacked clinical correlation and relied solely on a subjective angiographic measurement. These selection criteria likely created a cohort of widely variable clinical presentation which may not have been equitably distributed between the arms of the study. Although Vigen eta!. used weights to adjust for these imbalances, the two cohorts may not have been balanced with respect to underlying cardiac risk potential, and differences in cardiac outcomes may be due to this imbalance. In addition, the use of a composite outcome makes the interpretation of the study results less clear. Despite the data showing no significant difference between the testosterone and no* testosterone groups in the incidence of MI, stroke, and death, the authors' conclusions were contradictory to the crude event rates and were based on a complex statistical model using a weighting scheme involving approximately 50 variables. At least one key variable was not accounted for: the significantly lower baseline testosterone level in the testosterone group, which was viewed by the authors' peers as a significant oversight. 33 Finally, the exclusion of 128 patients who experienced MI or stroke before initiating testosterone was not appropriate. These patients should have been included in the analysis and their events included in the no-testosterone group, which would have raised the event rate in the no* testosterone group by 71%. Their exclusion biased the results by reducing the number of events in the no-testosterone group. This issue was also raised by other researchers in their comments regarding this article.3 The authors issued a correction stating that, while their original publication noted that 1,132 patients had been excluded, this was "an incorrect notation" and only 128 patients were excluded for having an MI before testosterone therapy was initiated.35 Although the authors claimed that including the 128 patients did not change the result (HR 1.3; 95% CI 1.1-1.6), their exclusion raises significant concerns about the quality of the study. It is also unclear why the authors excluded 1,301 participants for not having coronary anatomy data (CAD status), considering the wealth of baseline information collected on medical and drug history. It is unclear how this exclusion might also have affected the risk estimate. Given the described limitations of the study by Vigen et al. it is difficult to attribute the reported findings to testosterone treatment. [U]Finkle et al[/U] Finally, the Petition relies on the outcome of a 2014 study by Finkle eta!. The Petition claims that "the statistically significant findings were that the risk of heart attacks after using the drug [testosterone] for three months was twice the risk in the year before use in all men 65 and over. Furthermore, the study found for the first time that in those men under 65 with a history of heart disease, there was 2.9-fold increase in heart attack risk."36 Finkle conducted a retrospective cohort study to assess a possible association between testosterone therapy and non-fatal MI 90 days following an initial prescription. The Finkle study had the largest sample size of the four studies reviewed. Finkle used the MarketScan database to identify 55,593 patients prescribed testosterone and 167,279 patients prescribed a phosphodiesterase-S inhibitor (PDESI). The authors first used a self-control cohort method to compare the post-exposure incidence rate with the pre-exposure rate among the testosterone exposed cohort, and then used a parallel cohort method to compare the incidence rates between testosterone and PDESI patients. The results from the self-control analysis showed that for 55,593 patients initiating testosterone, there was a significant increased risk within the first 90 days following initiation of testosterone compared with the risk in one year preceding testosterone therapy (relative risk (RR) 1.36; 95% CI 1.03, 1.81). The subgroup analysis of age and history of heart disease in the testosterone patients also showed significant increased risks for post- versus pre-exposure for participants older than 65 years of age with no heart disease (RR 2.21; 95% CI 1.09-4.46), and less than 65 years of age with heart disease (RR 2.90; 95% CI 1.49-5.62). The results from the parallel cohort analysis were similar. Males in the testosterone cohort had an increased risk for non-fatal MI, ratio of the rate ratios (RRR) (1.27; 95% CI 0.94- 1.71). There was an approximate doubling of the risk for males older than 65-years old with heart disease (RRR 1.90; 95% CI 0.66-5.50) and without heart disease (RRR 2.41; 95% CI 1.12- 5.17), and for males younger than 65-years old with heart disease (RRR 2.07; 95% CI 1.05- 4.11). These data suggested a significant risk of MI in all patients prescribed testosterone, driven by those patients more than 65 years of age and those patients less than 65 years of age with a history of heart disease. For those patients older than 65 years of age with a history of heart disease taking testosterone, there was no significant risk for MI. Despite these results, the study has some limitations that raise questions about whether there is a true risk for non-fatal MI with testosterone therapy. The large size of the MarketScan administrative database allowed the investigators to evaluate the risk among a large number of patients. However, in these data, the diagnostic indications for testosterone were not available. Further, results of laboratory testing of testosterone levels are not available; this may be important as low serum testosterone is a known risk for cardiovascular events. Testosterone exposure was determined based on a patient's filling of a prescription for testosterone therapy, but it is unknown whether the patient actually used the prescription. This fact, combined with the inability to assess baseline or post-treatment testosterone levels or indication for therapy associated with the males treated with testosterone, makes it impossible to determine if the testosterone levels in these treated males had reached therapeutic range. Due to these limitations, it is difficult to completely attribute the increased risk for non-fatal MI to testosterone treatment. It is also questionable as to whether the self-controlled cohort or the active comparator design is most appropriate for assessing outcomes of testosterone therapy. Normally, testosterone is prescribed for chronic use, but Finkle limited follow-up to 3-months of therapy. It is unclear if 3-months' follow-up is adequate to capture the relevant outcomes. In the testosterone cohort, the males tended to be younger with a higher comorbidity load compared with males in the PDE5I cohort. For both the crossover analyses and the active comparator analysis, the overall risk was small. However, age and heart disease status appear to be confounding factors. In both the crossover and active comparator analyses, the authors found a 2-fold increased risk in males over 65-years old without heart disease and a 2-3 fold increased risk in males with heart disease regardless of age. In addition, acute non-fatal MI was the only outcome measured. The study subject would have had to survive to be included in the analysis. As fatal MI and other outcomes such as cardiovascular mortality or stroke were not captured, it is unclear how their inclusion would have affected the study results. Due to these uncertainties, it is difficult to attribute the increased risk for non-fatal MI seen in the Finkle study to testosterone alone and not consider that the study participants might have remained hypogonadic and thus at higher risk for non-fatal MI. [U]Other Literature[/U] In addition to reviewing the studies cited in the Petition and the literature search discussed above, FDA also performed a literature search to identify other articles that may be relevant to the question of whether testosterone can be linked to increased cardiovascular risk. We identified two relevant studies, which show either an apparent benefit of treatment with testosterone or an inference that testosterone therapy is not associated with an increased cardiovascular risk. The first, a 2012 study by Shores et a!.,37 is an observational study designed to examine the association between testosterone treatment and mortality in men with low testosterone. The database included seven Northwest Veterans Affairs medical centers and included a cohort of 1,031 male veterans older than 40 years of age with low testosterone (<250ng/dL). In this study, testosterone treatment was associated with a decreased mortality compared with no testosterone treatment (HR 0.61; 95% CI 0.42-0.88). Similarly, in a 2013 prospective follow-up from a previously reported cohort that collected data from outpatient medical facilities with access to medical records, Muraleedharan et a!.38 concluded that low testosterone levels predicted an increase in all-cause mortality during long* term follow-up, and testosterone replacement may improve survival in hypogonadal men with type 2 Diabetes Mellitus. The Shores study used a lower threshold for low testosterone (<250 ng/dL) to increase the likelihood of a symptomatic hypogonadal population. In both studies, lower testosterone level was associated with testosterone therapy, but Body Mass Index and younger age were also predictors of use of testosterone in the Shores study. Otherwise, comorbid conditions were balanced between the cohorts in each of the studies. Both studies used time-to-event Cox regression analysis to calculate the risk estimates and confidence intervals, and they each showed an approximate 50% reduction in the risk for death with testosterone therapy. In addition, the Shores study showed that increasing mortality is associated with lower baseline testosterone levels and shorter duration of testosterone therapy. This might indicate that sicker, undertreated males are at a higher risk for mortality. The Muraleedharan study showed a similar mortality rate for the diabetic males on testosterone and for a cohort of diabetic males who had normal testosterone levels and no testosterone therapy, which suggests the untreated males in this study have similar outcomes to the treated hypogonadal males. On average, duration of testosterone therapy was longer than two years for the majority of the treated subjects, and the peak testosterone level was 657 ng/dL. Over 67% of the treated males had a testosterone level of over 518 ng/dL. These levels are well within the recommended mid-to-normal therapeutic levels (400 -700 ng/dL). In both studies, only all-cause mortality was reported as the outcome, so outcomes of interest that did not result in death were not captured. [U]FDA's Response[/U] As discussed in this response, the studies presented in the Petition have significant limitations that weaken their evidentiary value for confirming a causal relationship between testosterone and adverse cardiovascular outcomes. These weaknesses include: • Short follow-up times precluding assessment of the potential for long term benefits of testosterone therapy (Finkle); • Unclear statistical methods (Vigen); • Inability to compare results across studies due to differing outcomes and populations (Vigen, Finkle); • Overall effect estimates are small and may be due, in part, to residual confounding (Vigen, Xu, Finkle); • Limitations with respect to ascertainment of events (Basaria, Finkle); • Overly broad case definition for cardiovascular events (Xu); • Incomplete or unavailable laboratory data to confirm hypogonadism or to assess whether patients returned to normal testosterone levels after receiving treatment (Finkle, Vigen); • Failure or inability to assess other potentially relevant laboratory data such as hematocrit or hemoglobin (Finkle); • Non-validated endpoints or lack of compliance data (Finkle, Vigen, Xu); and • Conflicted results suggesting both a testosterone benefit (Shores and Muraleedharan) and testosterone harm with respect to cardiovascular risk, or no difference between groups (Vigen, Xu). In addition, FDA has identified other studies in the literature that contradict the findings in the studies submitted. Prior to the submission of the Petition, the Agency had already undertaken a thorough evaluation of the literature and other evidence to determine if additional regulatory action is necessary to protect consumers from the cardiovascular risks of testosterone therapy. As our January 31, 2014, drug safety communication indicated, FDA believes that the publication of these studies warrants further exploration of a possible safety signal regarding testosterone and cardiovascular risk. Our current evaluation remains ongoing. For the reasons discussed above, the Agency does not believe at this time that the evidence presented in the Petition is sufficient to require the addition of a boxed warning regarding cardiovascular risks of testosterone therapy to the labeling of all testosterone products. Therefore, at this time, FDA declines to exercise its authority to require safety labeling changes regarding these risks on the basis of the evidence presented in the Petition, and your request is denied. Consequently, your requests that FDA require FDA* approved Medication Guides for testosterone products to be updated with the same warnings and that manUfacturers be required to send Dear Doctor Letters regarding these risks are also denied. We are continuing to assess this potential safety signal. In particular, we are awaiting the results of the Testosterone Trial,39 a multicenter study of six coordinated trials investigating the effects of testosterone treatment in elderly men with low testosterone on physical function, vitality, sexual function, cognitive function, anemia, and cardiovascular risk. Eight hundred men over 65-years-of age whose serum testosterone is less than 250 ng/dL have been or will be randomized to receive testosterone or placebo double blindly for one year. Although this trial is not a safety study, we believe that the data will yield important information regarding the safety of testosterone with regard to cardiovascular risks. In addition, we intend to present the question of the potential association between testosterone and adverse cardiovascular events to an Advisory Committee this fall. Based on the outcome of these efforts, FDA intends to make a determination as to whether any regulatory action is warranted, such as invoking our authority to require safety labeling changes under section 505(o)(4) of the FD&C Act for testosterone-containing drugs, as appropriate. [B]III. CONCLUSION[/B] After careful consideration, and, in light of the foregoing, we hereby deny your Petition in its entirety. FDA will continue to evaluate the cardiovascular risks of testosterone, and, if warranted, will take appropriate regulatory action to protect the public health when its evaluation has concluded. Janet Wood****, M.D. Director Center for Drug Evaluation and Research 29 Protocol: Testosterone and cardiovasclar related events in men: a meta.analysis of randomized controlled trials. 12-5-2011. [URL]http://www.crd.york.ac.uk/PROSPEROFILES/1815[/URL] [U]PROTOCOL 20111108.pdf.[/U] 30 Petition at 4. 31 Incorrect Number of Excluded Patients Reported in the Text and Figure. [I]Journal of American Medical Association [/I]2014, 311: 967. (Incorrect Number); Comment and Response. [I]Journal of American Medical Association [/I]2014, 311: 964-965. (Comment and Response).32 The Endocrine Society Clinical Practice guidelines recommend raising serum levels to between 400 and 700 ng/dL. See Bhasin S, Cunningham GR, Hayes FJ et al. Testosterone therapy in men with androgen deficiency syndromes: an Endocrine Society clinical practice guideline. [I]J [/I][I]C/in Endocrinol Metab [/I]2010;95:2536-2559. 33 Traish, A, eta!., 2014, Death by testosterone? We think not! [I]J [/I][I]Sex Med, [/I]11:624-629. 34 Comment and Response, 2014. 35 Incorrect Number, 2014.36 Petition at 5.37 Shores, M, et al., 2012, Testosterone treatment and mortality in men with low testosterone levels, [I]J [/I][I]Clinical[/I] [I]Endocrinol Metab, [/I]97 (6):2050-2058. 38 Muraleedharan, V, et al., 2013, Testosterone deficiency is associated with increased risk of mortality and testosterone replacement improves survival in men with type 2 diabetes, [I]Eur [/I][I]J [/I][I]Endocrino/, [/I]169: 725-733. 39 For more information regarding the Testosterone Trial, see [URL="http://www.med.upenn.edu/idom/t-trial.html"]http://www.med.upenn.edu/idom/t-trial.html.[/URL] [/QUOTE]
Insert quotes…
Verification
Post reply
Share this page
Facebook
Twitter
Reddit
Pinterest
Tumblr
WhatsApp
Email
Share
Link
Sponsors
Forums
Testosterone Replacement, Low T, HCG, & Beyond
Testosterone and Men's Health Articles
F.D.A. Panel Backs Limits on Testosterone Drugs but Rejects Petition for More Regulations
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
Accept
Learn more…
Top