Reports of Increased Mortality and GH: Will This Affect Current Clinical Practice?

Although randomized controlled studies are considered the “gold standard” to assess efficacy and safety, this approach is not always practical to ascertain long-term safety. To better define emerging safety signals, particularly for products already on the market, we mostly rely on observational studies. However, we are cognizant that all observational studies suffer from a critical deficiency: the design is not an experimental one. Because each patient's treatment is deliberately chosen rather than randomly assigned, the risk of selection bias is unavoidable. As a result, systematic differences in outcomes may not be due to the treatment itself. Although methods to adjust for identifiable differences are available, it is impossible to be certain that such adjustments are sufficient or whether they address all patients' relevant characteristics. Estimates of the magnitude of the treatment effect and on the generalizability of the findings are therefore imprecise. All these caveats apply to any observational study. Observational studies do not definitively answer a question, but rather generate hypotheses that need to be further explored with additional studies.

Questions regarding the short- and long-term safety of GH have been gaining attention due to the expansion of GH indications and the use of increased dosages. In this issue of the JCEM, Carel et al. (1) report the results of a longitudinal follow-up study of subjects receiving GH where they ascertained its long-term effects on mortality. By 2009, after a mean follow-up of 17.3 yr, they were able to determine the vital status of about 95% of approximately 7000 subjects who received GH in France. Indications for GH varied, but 75% were labeled as idiopathic GH deficient (IGHD), 11.5% as idiopathic short stature, 8% as having a GH neurosecretory disorder, and 5.5% as small for gestational age. Ninety-three subjects died. Eighty-six deaths occurred after GH discontinuation, but six occurred while on active treatment. All-cause mortality was 33% higher than expected when compared with the general population. This increase was attributed to a higher than expected number of bone tumors, as well as subarachnoid and intracerebral hemorrhages. The increases were not associated with any other types of cancer. Doses in excess of 50 μg/kg·d were strongly associated with the increase in mortality. Significance was seen only after 15 yr of follow-up. Analyses at 5 and 10 yr did not show a significant increase in mortality compared with the general population. Shorter children at baseline and males were at increased risk, as well as those with more robust responses to provocative tests.

The strengths of this study (1) include the balanced review and fair presentation of the information; the extent of the data, including diagnoses, dosages, time of exposure, and long-term follow-up; the attempts to ascertain in depth causes of death; and the use of internal and external controls.

A major weakness of this study (1) is the lack of efficacy data. We do not know how these subjects responded to the treatment. Other limitations are related to the inability to identify and secure an appropriate short stature comparator, the lack of data on growth velocity before GH was administered, the absence of IGF-I levels before and during GH, the small number of events, and the significant amount of missing data, particularly in the ascertainment of cause of death in 22% of cases. The mean age at the time of censoring was less than 30 yr, and it is unknown whether the encountered trend will continue or disappear with passage of time. The number of females may be too small to detect any effect in this group. Many of these shortcomings were recognized by the authors, and attempts to address them were made. Each of these deficiencies, however, opens the door to alternative explanations. It is important to stress that the difficulties facing these authors are common in the field, and that investigators undertaking this sort of project are likely to encounter similar barriers.

Relevant Variables for Analyses of These Data (1)

Dosage

Dosing is probably one of the most reliable variables in this study (1). Although errors may occur when prescribing and administering medications, in general, dosage data are likely to be robust. The results of this work suggest that for every 75 individuals treated with GH, one will experience an earlier than expected death. However, this increased risk is not equally present for all subjects. Clustering of premature deaths within the subgroup that received higher GH dosages is clear. This risk may be up to three times greater than for those receiving this currently approved dose for GH replacement.

Is there a known association between GH treatment and increased risk of mortality? One prior study demonstrated that intensive-care patients treated with GH had an increased risk of mortality (2). In addition, a temporal relationship has been observed between the initiation of GH treatment in pediatric patients with Prader-Willi syndrome and premature mortality (3). These two publications (2, 3) describe premature mortality at the time of, or soon after, GH treatment, whereas the manuscript of Carel et al. (1) reports premature mortality occurring years after the discontinuation of GH treatment.

There are also precedents from well-designed and large randomized controlled studies where morbidity and mortality were associated to aggressive treatment. Attempts focusing on the normalization of markers such as hemoglobin A1c (4, 5) and hemoglobin (6, 7) have also shown increased mortality associated with treatment related to higher dosages of diabetes medication in patients with type 2 diabetes and erythropoietin in anemia secondary to chronic kidney disease. In this context, the use of higher doses of GH is concerning.

Diagnosis

Although the data set was incomplete, 71% of subjects followed in this cohort had heights greater than the third percentile, placing them in the normal height range. In addition, 16% had heights falling within 2 sd values below the mean. Given this information and considering that the mean age was approximately 11 yr, it is highly unlikely that 75% fit the definition of IGHD. Unless all of these children had parents with heights greater than the 50th percentile, which is very unlikely, a child with IGHD would be expected to deaccelerate growth and consistently continue to fall down into lower height percentiles during infancy. However, most of the individuals in this cohort had heights within the normal range, challenging the IGHD diagnosis.

Provocative tests

Sixty-nine percent of patients had GH responses in excess of 7 mg/liter. GH response to GH provocative testing was greater than 10 mg/liter in 23% of treated children. The overlap between anthropometric parameters and these GH responses is not clear. The mean age was 11 ± 3.4 yr. There is no information as to the test used to ascertain GH responses or whether sex steroid priming was administered in those meriting this intervention. Also, during this period of follow-up, the different assays used to measure GH changed. Therefore, GH values cannot be easily compared without in-depth analyses of each method used. In addition, given the peripubertal age of the cohort, if sex steroid priming had been used on the appropriate candidates, the percentage of responders (those with stimulated GH levels > 10 mg/liter) would have been probably greater.

Cancer and mortality

The investigators reject the hypothesis that cancer is increased in patients previously treated with GH (8, 9). However, this cohort seems to differ from those previously reported. This population seems to be healthy. The target population is a critical element when assessing risks and benefits. Even at similar dosages, different populations may have distinctive efficacy and safety responses. The best example of differential responses of GH are the distinct and unique safety profiles observed between pediatric and adult GH deficiency (GHD) patients. In contrast to adults, children tolerate much higher GH dosages.

This study (1) fails to replicate what was previously published and brings to the forefront other potential causes of death that were previously unrecognized. Because the data in this study (1) are quite robust, one potential conclusion is that what was previously reported is not substantiated by these data and that the association between cancer and GH, albeit weak, is a fluke. This contention is further supported by a similarly designed study, albeit not identical, that did not demonstrate an association between GH treatment and cancer and, in addition, failed to show increased mortality (10). Deaths attributed to bone tumors and cardiovascular events were not reported. Suicide and fatalities associated with accidents were the principal causes of death. Hence, it is possible that the findings reported here, in contrast to speculation by the authors, do not have an identifiable biological basis and were random and/or pertain only to this French cohort. Whether this patient population is comparable to those previously published is also questionable, due to the many differences between patient populations, diagnostic criteria, dosages, and time exposure.

Biological explanations

Because we tend to look for biologically plausible hypotheses, it is unlikely that the complications seen in acromegaly could explain any untoward findings with these pharmacological doses of GH. Acromegaly (11) is a condition resulting from sustained GH excess. Few subjects in this cohort received GH dosages that would result in GH levels comparable to those seen in acromegaly. Individuals with acromegaly have severe GH excess and numerous concomitant metabolic derangements for years. In the current study, the average treatment exposure was less than 4 yr and even less for those at greater risk, whereas patients with acromegaly present after many years of GH excess. Any metabolic effects due to high GH would have resolved days after GH was discontinued. In contrast, normalization of metabolic variables is not always achievable in a significant number of patients with acromegaly. Moreover, bone tumors are not common in patients with acromegaly, and the cardiovascular complications described in acromegaly are not consistent with the cerebrovascular findings in this study (1). Finally, the patients treated with GH were children, and not adults, at the time of treatment, and it is possible that different effects and complications might result from treatment during childhood, compared with adults with even similar exposure.

Efficacy

This information is not provided. We do not know how much these subjects grew or the final heights they attained. These outcomes may have been quite different by diagnostic group or dose. We do not know whether the group identified as having more risk responded to treatment and how these responses compare to the other patients. Without this information, it is impossible to perform a risk-benefit analysis. Comparisons with available efficacy data obtained in other studies could be considered, but this will be highly problematic.

Summary

As in the past, when reports emerged associating GH with increased risk of cancer, the information necessary to identify those at increased risk is limited. However, several issues emerge from this study (1). They should be considered in light of the rejection of a principal rationale for GH intervention in short children, namely the psychological disturbances associated with extreme short stature (12) and the lack of data on efficacy.

Surprisingly, more than 70% of the children in this study (1) had normal stature. This fact causes one to wonder whether these findings apply to truly GHD patients who have significant short stature fail to respond to provocative tests and receive doses significantly lower than those that were associated with this negative outcome. These results pertain to children with normal stature as well as children with extreme short stature who respond appropriately to GH provocative testing. This latter group, which received the highest doses of GH, was more prone to experience premature death compared with what would be expected after discontinuation of GH. Unfortunately, we do not know how well they responded, if they responded at all. Thus, there is an emerging parallel with previous reports on GH and increased mortality risk (2, 3). When using higher dosages, untoward outcomes emerge. In fact, this information suggests that GH-resistant patients should not receive higher doses of GH (2), as also observed with erythropoietin and antidiabetic drugs (5, 6).

Prescription of GH to most individuals with normal stature and normal GH provocative testing is medically questionable. If patients are treated in the absence of support from anthropometric data and laboratory analyses, the burden of risks of treatment appears to outweigh the benefits.

When making therapeutic decisions, the balance between the strength of the desired effects contrasted with the known risks allows for more rational therapeutic choices. A compound that is marginally effective and poses high risks may be very useful in the absence of other therapeutic options or in devastating or life-threatening illnesses. In contrast, a compound that is devoid of safety concerns, if not effective, does not present advantages.

This study (1) suggests that the community is overdiagnosing GHD by mislabeling children with normal stature without considering the value of growth velocity and provocative testing as linchpins for deciding who and when to treat and at what doses. This study (1) also suggests that those with less potential for benefit are at greater risk of death after discontinuation of GH. Of concern is that, despite the lack of evidence of benefit or safety, higher dosages are being used and these are becoming the new standard of care.

Again, the lack of information assessing responses to therapy limits the ability to put these data into perspective. With the data presented, it is not possible to speculate on what is the inflection point benefit that may trigger the decision to prescribe GH in a particular patient, given the risk reported.

Conclusions

Carel et al. (1) provide information suggesting that high dosages of GH, particularly in short children with heights below the third percentile and that responded with adequate GH elevations to GH provocative tests may be at increased risk of premature death years after GH is discontinued. The observational nature of the study cannot, however, establish accurately that this is correct. Lack of replication of any of these findings by a similar study under review further raises questions about the link between GH and long-term mortality risk (10). The use of GH in normal children is problematic, and the use of high doses more so.

Independently of these findings, I do not foresee a change in prescribing practices as a result of this report (1). Changes in behavior rely on substantial incentives. Unfortunately, data and results alone seem less important than strongly held beliefs. Few pediatric endocrinologists treat a large number of subjects, and it is extremely unlikely that a death will occur while on treatment. In addition, pediatric endocrinologists do not follow patients once they enter adulthood, making it highly improbable that they will become aware of deaths in patients previously exposed to GH. Hence, the incentive to change currently adopted prescribing practices is weak, despite any concerns that this study (1) may elicit.