A comparative quality assessment of evidence-based clinical guidelines in endocrinology.

CONTEXT: Evidence-based clinical guidelines in endocrinology attempt to improve and standardize patient care. There has been an expansion in guideline production although some of the heterogeneous methods used to assess the quality of the underlying evidence base might limit interpretation and implementation. DESIGN: Current and archived guidelines from major endocrine organizations were accessed. The organizations used six different methods to rate underlying evidence, including Grading of Recommendations Assessment, Development and Evaluation (GRADE). To allow direct comparison between guidelines produced by different organizations, the levels of evidence used to generate them were graded according to the standardized system: 'high' based on randomized, controlled trials and meta-analyses, 'moderate' based on nonrandomized studies and 'low' based on expert opinion. RESULTS: There was an increase in guideline production over time (1995-2000 = 9, 2001-2005 = 12, 2006-2011 = 36). Three guidelines were updated with an average delay of 4·3 years and an increase in recommendations per guideline (21·1%). Encouragingly, whilst updates had similar levels of 'high'-quality evidence, there was increased reliance on 'moderate'-category evidence and less on 'low''-quality evidence' ('high', 6·3% vs 6·5%; 'moderate', 46·1% vs 59·1%; 'low', 47·7% vs 34·4%). A high proportion of 'low'-category evidence was seen throughout all organizations. Rarer conditions and recommendations concerning treatment efficacy were particularly reliant on 'low'-category evidence. CONCLUSIONS: The level of evidence underpinning current guidelines highlights areas in need of well-designed, collaborative clinical research. Furthermore, criteria to define when guideline updates are necessary are currently lacking. A standardized method of assessment, such as GRADE, would promote understanding and compliance by guideline users with the ultimate aim of enhancing patient care.

Mitotane therapy in adrenocortical cancer induces CYP3A4 and inhibits 5α-reductase, explaining the need for personalized glucocorticoid and androgen replacement.

CONTEXT: Mitotane [1-(2-chlorophenyl)-1-(4-chlorophenyl)-2,2-dichloroethane] is the first-line treatment for metastatic adrenocortical carcinoma (ACC) and is also regularly used in the adjuvant setting after presumed complete removal of the primary tumor. Mitotane is considered an adrenolytic substance, but there is limited information on distinct effects on steroidogenesis. However, adrenal insufficiency and male hypogonadism are widely recognized side effects of mitotane treatment. OBJECTIVE: Our objective was to define the impact of mitotane treatment on in vivo steroidogenesis in patients with ACC. SETTING AND DESIGN: At seven European specialist referral centers for adrenal tumors, we analyzed 24-h urine samples (n = 127) collected from patients with ACC before and during mitotane therapy in the adjuvant setting (n = 23) or for metastatic ACC (n = 104). Urinary steroid metabolite excretion was profiled by gas chromatography/mass spectrometry in comparison with healthy controls (n = 88). RESULTS: We found a sharp increase in the excretion of 6ß-hydroxycortisol over cortisol (P < 0.001), indicative of a strong induction of the major drug-metabolizing enzyme cytochrome P450 3A4. The contribution of 6ß-hydroxycortisol to total glucocorticoid metabolites increased from 2% (median, interquartile range 1-4%) to 56% (39-71%) during mitotane treatment. Furthermore, we documented strong inhibition of systemic 5α-reductase activity, indicated by a significant decrease in 5α-reduced steroids, including 5α-tetrahydrocortisol, 5α-tetrahydrocorticosterone, and androsterone (all P < 0.001). The degree of inhibition was similar to that in patients with inactivating 5α-reductase type 2 mutations (n = 23) and patients receiving finasteride (n = 5), but cluster analysis of steroid data revealed a pattern of inhibition distinct from these two groups. Longitudinal data showed rapid onset and long-lasting duration of the observed effects. CONCLUSIONS: Cytochrome P450 3A4 induction by mitotane results in rapid inactivation of more than 50% of administered hydrocortisone, explaining the need for doubling hydrocortisone replacement in mitotane-treated patients. Strong inhibition of 5α-reductase activity is in line with the clinical observation of relative inefficiency of testosterone replacement in mitotane-treated men, calling for replacement by 5α-reduced androgens.