The Effect of Growth Hormone on Life Expectancy
Is Cancer An Issue?
Feature Article - e-Journal of Age Management Medicine - December 2006
By Alan P. Mintz, M.D.
Life Expectancy
Prior to the 1980s, panhypopituitary patients rarely received human Growth Hormone (hGH). Stavrou confirmed the significantly decreased life expectancy in hypopituitary patients with GH deficiency (GHD), with cardiovascular disease a common cause of death. He also noted that treatment with hGH reversed abnormalities in body composition which may reduce cardiovascular risk factors.
In 1990, Rosen reviewed 333 consecutive patients with hypopituitarism diagnosed between 1956 and 1987. All of these patients had received routine hormonal replacement therapy, save hGH. Yet, overall mortality was higher for these patients than their age- and sex-matched cohorts. He concluded that life expectancy is shortened in patients with hypopituitarism and that GHD could be a factor in increased mortality from cardiovascular disease.
Bulow performed a retrospective cohort study from 1952 to 1992 of 344 patients who received conventional hormone replacement (again, save for hGH) for hypopituitarism following neurosurgical intervention for pituitary tumors and found an increased mortality from cerebrovascular and cardiovascular disease.
Tomlinson followed 1014 UK patients with hypopituitarism from January, 1992 to January, 2000 and noted a profound excess mortality, predominantly from cardiovascular, cerebrovascular and respiratory diseases. Multiple regression analyses identified age at diagnosis, sex, craniopharyngioma and untreated GHD as independent significant factors affecting mortality.
Colao, in a short study, followed 30 GHD patients and 30 healthy subjects for 1 year. 15 GHD patients agreed to hGH therapy while the other 15 did not. 12 short months of hGH therapy normalized IGF-1, improved lipid profiles and increased cardiac performance in adult GHD patients while a similar period of (voluntary) hGH deprivation induced further impairment of lipid profile and cardiac performance.
Besson, in a very elegant study, compared 11 subjects with hereditary dwarfism and a genetic defect leading to isolated GHD, to their 25 unaffected siblings as well as the normal population. He found that the median life span in the GHD group was significantly shorter than that of the unaffected siblings, 19 years less for males and 34 years less for females. He concluded, rightly so, that hGH treatment in adult patients suffering from either childhood- or adult-onset GHD is crucially important.
As you well know from data obtained over this past decade, the short-term benefits of hGH supplementation for the GHD include improvements in energy, mood & body composition, specifically, decrease fat & increase muscle/strength & bone density, via Insulin-like Growth Factor-1 (IGF-1). However, controversy and major concerns exist about the "unknown", eg long-term risks, specifically cancer.
Physiology
Let's get some facts on the table up front. The IGF family includes IGF-1 and -2, at least 2 IGF Receptors (IGFR) and 6 IGF Binding Proteins (IGFBP), some of which are pro-apoptotic while others are not. IGF-1 is a known tissue mitogen and stimulates the growth of cancer cells in vitro. Essentially, IGF-1 is a potent inhibitor of apoptosis or programmed cell death. Most cells, including cancer cells, possess IGFR-1 and respond to IGF-1 with increased growth. Overexpression of IGFR-1, but not systemic administration of IGF-1, induces tumor formation in animal models. While no one has demonstrated the transformation of a normal cell line into a cancerous cell line solely as a result of exposure to IGF-1, systemic administration of IGF-1 can increase the growth of established tumors.
Epidemiology
Many epidemiologic studies have demonstrated an association between elevated levels of IGF-1 and cancers of various sorts. However, this is similar to stating that many studies have noted an association between leukocytosis and infections or pain and fractures. Clearly, no one would claim that high levels of white blood cells lead to infections or that pain leads to fractures, based upon these studies. Yet, we are ready to believe that elevated IGF-1 leads to cancer in the body even though cause and effect have never been proven.
While some epidemiologic studies have also demonstrated an association between high normal levels of IGF-1, low levels of IGFBP-3 and breast, colorectal and prostate cancer, others have not demonstrated any linkage. While stating that IGF-1 levels have not been statistically associated with cancer risk, Ma noted that high IGF-1 and low IGFBP-3 appear to be related to a heightened risk. A year later, Yu again noted that high levels of circulating IGF-1 and low levels of IGFGP-3 are associated with an increased risk of breast, colorectal, lung and prostate cancers. However, because hGH positively influences the production of both, it's doubtful that hGH acts as a driving force in the IGF-1 cancer relationship.
In analyzing KIMS, the Pharmacia & Upjohn International Metabolic Database, a long-term, open, outcomes research program of hypopituitary adult patients with GHD treated in a conventional clinical setting, Abs found 6 cases of recurrence of pituitary or central nervous system tumor, a rate consistent with data from control series and no increased incidence of any cancer in 1034 adults treated with hGH for a total of 818 patient treatment years. Further analysis of KIMS by Bengtsson revealed no excess mortality in 1903 patients treated for 2334 patient treatment years with mean treatment time of 1.2 years, maximum treatment of 4 years, 63% of the treatment time beyond 6 months, and 35% beyond 12 months.
Likewise, Isley, in a review, found no increase in rates of cancer in adult patients who had received hGH. In a small study of 24 GHD patients followed for 12 months, Colao found no prostate issues in those given hGH therapy plus or minus testosterone.
Acromegaly
Let's then consider those who've been labeled an experiment of nature, those with acromegaly or abnormally high levels of hGH production leading to elevated levels of both IGF-1 and IGFBP-3. Cook, in another review, noted that acromegalics have not been associated with an increase in cancer, suggesting that high-normal IGF-1 concentrations may be a marker for cancer but are not causally related to inducement or growth of cancer.
Renehan prospectively screened 122 acromegalics 25-82 years old by colonoscopy and found no association between colonic neoplasia and duration of disease, total hGH exposure, cure status and IGF-1. Due to lack of ideal age-matched controls, these patients were compared to published data from 8 autopsy studies (n = 3559) and 4 screening colonoscopy studies (n = 810). Essentially, there was no increase prevalence of neoplasia in acromegalics compared to the expected prevalence in the general population.
Over 5 years, Ladas colonoscoped 54 acromegalics and gastroscoped 42 but found no increased incidence of gastric tumors. Colao studied 46 acromegalics and found a higher rate of benign prostate hyperplasia but not of prostate cancer.
Melmed reviewed 12 prospective colonoscopy studies reporting on 678 patients with acromegaly. Approximately 47% of these patients were found to harbor either an adenoma, hyperplastic polyp or frank colon carcinoma, leading to the assumption that acromegaly is associated with colon cancer. However, Melmed himself stated that 40% of asymptomatic unselected males over age 50 in the general population harbor colonic lesions and that appropriate age-, sex-, and environmentally-matched control population groups were needed before extrapolating any data from his literature review. In fact, while cancer is the 2nd leading cause of death here in the US, Melmed found that neoplasia accounted for only 15% of deaths in acromegalics. He summarized by stating that there exists no direct proven causal relationship between acromegaly and malignancy.
Mustacchi's 12 center analysis of 2981 acromegalics up to 79 years old found no observed over expected increase in cancer prevalence and concluded that there is no definite influence of the pituitary on initiation of neoplasia. Orme, in the UK, found that in a multicenter retrospective cohort study of 1362 acromegalics followed for 19,323 patient years, overall cancer incidence rate was lower than in the general UK population.
Childhood Cancer Survivors
Lastly, let's consider the results of hGH supplementation in survivors of childhood cancer. Sklar found that neither death nor risk for recurrence was increased for any major cancer diagnosis in 361 childhood cancer survivors treated with hGH for a minimum of 5 years. Murray found no clinical suggestion of tumor recurrence during 12 months of hGH replacement. The Lawson Wilkins Pediatric Endocrine Society (LWPES) endorses only a 1 year wait after completion of tumor treatment without further evidence of tumor recurrence prior to starting hGH replacement in children.
While acknowledging the existence of early reports of an increase in childhood leukemia following hGH therapy, Sperling stated that these findings were unfounded. Upon further analysis of growing international databases involving more than 86,000 patients on hGH, representing almost 250,000 hGH patient treatment years, there was but one report of a gastrointestinal carcinoma and a personal communication of a spontaneous colon cancer. In other words, there is no increase in tumor recurrence in those whose primary lesion has been successfully treated. He closed by stating that current treatment recommendations with hGH for children with documented GH deficiency are safe and without proven increased risk of inducing malignancy. That said, because any person treated for a malignancy is at risk for a second malignancy, ongoing surveillance is recommended anyways.
Swerdlow followed a cohort of 1848 patients in the UK treated in childhood and early adulthood with hGH between 1959 and 1985 and followed through December, 1995 for cancer incidence and through December, 2000 for mortality, accounting for 29,817 person-years of follow-up for cancer incidence (average 16.1 years per patient) and 39,178 person-years of follow-up for mortality analysis (average 21.2 years per patient). The overall risk of cancer incidence in the cohort was not significantly greater than that of the general population. Even more important, there were no cases of leukemia or breast or prostate cancer. Swerdlow did find a 3-11 fold increase in mortality due to colorectal cancer and Hodgkin's disease. There was also an 8-fold increase in the incidence of colorectal cancer as well as an increase in the incidence of Hodgkin's disease. However, the statistics are rather skewed in that there were only 2 deaths each of colorectal cancer and Hodgkin's disease. Thus, the chance occurrence of one case or death from each would have altered the statistics rather drastically.
Leung followed a smaller cohort of patients treated for acute lymphoblastic leukemia (ALL) who then received hGH supplementation. At 7 and 11 years in continuous hematologic remission, there was no evidence that hGH lead to leukemia relapse or development of a second malignancy.
Summary
Thorne summarized the state of knowledge regarding hGH supplementation very nicely:
1) those with previous malignancies or history of radiation therapy carry a significant risk for recurrence and 2nd malignancy
2) overexpression of IGF-1 receptors induces tumor formation in animal models
3) no evidence exists that systemic administration of IGF-1 to animals stimulates tumor formation de novo, although it can increase growth of established tumors
4) hGH has been successfully used in more than 86,000 children
5) no evidence that hGH therapy increases tumor recurrence in those successfully treated for primary lesion
6) no evidence that hGH therapy increases de novo tumor formation
7) combination of high IGF-1 and low IGFBP-3 levels may be associated with increased risk of breast, colon and prostate cancers
8) high levels of IGFBP-3 alone have also been related to reduced cancer risk
9) hGH increases both IGF-1 and IGFBP-3 levels
Conclusion
Dr. Pinchas Cohen, Director, Pediatric Endocrinology, UCLA, was quoted in the September 22, 2003 issue of Newsweek, “All the evidence shows that growth hormone is one of the safest drugs we have. Thousands of patients have been followed for the last 16 years."
References
Stavrou S. Endocrinol Metab Clin North Am. 01-SEP-2001; 30(3):545-63
Rosén T, Bengtsson BA. Lancet. 04-AUG-1990; 336(8710):285-8
Bülow B. Clin Endocrinol (Oxf). 01-JAN-1997; 46(1):75-81
Tomlinson JW et al. Lancet. 10-FEB-2001; 357(9254):425-31
Colao A. J Clin Endocrinol Metab. 01-MAR-2002; 87(3):1088-93
Besson A. J Clin Endocrinol Metab. 01-AUG-2003; 88(8):3664-7
Ma J, Pollak MN, Giovannucci E, et al. J Natl Cancer Inst. 07-APR-1999; 91(7):620–5
Yu H & Rohan T. J Natl Cancer Inst. 20-SEP-2000; 92(18):1472-89
Abs R et al. Clin Endocrinol (Oxf). 01-JUN-1999; 50(6):703–13
Bengtsson BA - J Clin Endocrinol Metab - 01-NOV-1999; 84(11):4291-2
Isley WL. Ann Intern Med. 06-AUG-2002; 137(3):190-6
Colao A. J Clin Endocrinol Metab. 01-JAN-2003; 88(1):88-94
Cook DM. Ann Intern Med. 06-AUG-2002; 137(3):197-201
Renehan AG et al. J Clin Endocrinol Metab. 01-SEP-2000; 85(9):3417-24
Ladas SD. Clin Endocrinol (Oxf). 01-NOV-1994; 41(5):597-601
Colao M. J Clin Endocrinol Metab. 01-JUN-1999; 84(6):1986-91
Melmed S. J Clin Endocrinol Metab. 01-JUL-2001; 86(7):2929-34
Mustacchi P & Shimkin MB. Cancer. JAN-1957; 10:100-104
Orme SM. J Clin Endocrinol Metab. 01-AUG-1998; 83(8):2730-4
Sklar CA. J Clin Endocrinol Metab. 01-JUL-2002; 87(7):3136-41
Murray RD. J Clin Endocrinol Metab. 01-JAN-2002; 87(1):129-35
Wilson TA. J Pediatr. 01-OCT-2003; 143(4):415-21
Sperling MA. J Clin Endocrinol Metab. 01-DEC-2002; 87(12):5351-2
Swerdlow AJ. Lancet 27-JUL-2002; 360(9329):273–7
Leung W. J Clin Oncol. 1-JUL-2002; 20(13):2959-64
Thorner MO. J Clin Endocrinol Metab. 01-MAY-2001; 86(5):1868-70
Alan P. Mintz, M.D., is founder and CEO of Cenegenics Institute.
|
