Assessment of disease activity in treated acromegalic patients using a sensitive GH assay: should we achieve strict normal GH levels for a biochemical cure?

The definition of a cure for acromegaly is controversial in the absence of a well-defined clinical end-point. Therefore, cure in acromegaly may be arbitrarily defined as a normalization of biochemical parameters. The accepted normal GH levels have been modified over time with the improved sensitivity of GH assays. The objective of the present study was to investigate the suppression of GH levels in the oral glucose tolerance test (oGTT) using a sensitive GH immunoassay in a large group of normal adult subjects and treated acromegalic patients. We evaluated these results in conjunction with IGF-I and IGF binding protein 3 (IGFBP-3) levels. Nadir GH levels after the ingestion of 75 g of glucose, as well as baseline IGF-I and IGFBP-3 levels, were evaluated in 56 normal adult subjects and 32 previously treated acromegalic patients. GH was assayed by an immunofluorometric assay. Normal controls had a mean GH nadir of 0.07 +/- 0.09 microg/liter. Their mean basal IGF-I and IGFBP-3 levels were 160 +/- 58 microg/liter and 1926 +/- 497 microg/liter, respectively. Acromegalic patients had mean GH nadir, IGF-I, and IGFBP-3 levels higher than those of normal subjects (2.6 +/- 7.6 microg/liter, 313 +/- 246 microg/liter, and 2625 +/- 1154 microg/liter, respectively). Considering a GH cut-off value of 0.25 microg/liter, as the normalized postglucose GH upper limit (mean + 2 SD) and, therefore, the target for treated patients, only five patients (15.6%) would have been considered cured. These results suggest that the strict physiological normalization of GH levels after oGTT is not often achieved as a therapeutic endpoint in acromegaly. In addition to the refinement of GH assays, epidemiological studies have suggested that the mean basal GH levels (<2.5 microg/liter) or oGTT-derived GH levels < 2 microg/liter (RIA), or the normalization of IGF-I levels, appear to reduce morbidity and mortality in treated acromegaly. Using this epidemiologically based definition of cure for acromegaly, we reviewed our results obtained with a sensitive GH assay. Twenty-five patients (78%) had oGTT nadir GH < 2 microg/liter. Nineteen subjects had normal age-related IGF-I levels. When the GH nadir cut-off was reduced to 1 microg/liter or less, there was a cure rate of 59.4%. IGF-I and IGFBP-3 levels were normal in 16 and 15 of these 19 patients, respectively. Furthermore, 59.4% of these 32 patients were in remission when age-normalized IGF-I levels were used as a criterion for inactive disease. All but three had GH nadir of 1 microg/liter or less. Finally, the definition of cure may be contradictory in a subgroup (9.4%) of patients with a GH nadir less than 1 microg/liter despite high-for-age IGF-I levels. In conclusion, using a sensitive GH assay it can be seen that the strictly normal postglucose GH values less than 0.25 microg/liter required for biochemical control of acromegaly are not often achieved. Furthermore, the cut-off of GH nadir 1 microg/liter or less is more closely related to normal for age serum IGF-I levels in treated acromegalic patients than 0.25 microg/liter or 2 microg/liter cut-offs. According to previous epidemiological reports, a GH level less than 2.5 microg/liter, determined by RIA, is associated with a reduction of morbidity and mortality. Therefore, our data lead us to postulate that the biochemical criterion of oGTT GH levels 1 microg/liter or less, determined by immunofluorometric assay, is a useful and accurate marker of safe GH secretion in treated acromegaly.