Growth hormone deficiency in children treated for leukaemia

Growth hormone treatment and cancer risk

Increasing numbers of children receive growth hormone (GH) to treat a range of growth disorders, including those rendered GH deficient (GHD) by tumors or their treatment. Young persons with persistent growth hormone deficiency (GHD) and adults with severe GHD are also eligible to receive GH treatment. As in vitro and in vivo studies and epidemiologic observations provide some evidence that the GH--insulin like growth factor-I (IGF-I) axis is associated with tumorigenesis, it is important to assess, in practice, the incidence of tumors related to GH treatment. Reassuringly, surveillance studies in large cohorts of children and in smaller cohorts of adults indicate that GH is not associated with an increased incidence of tumor occurrence or recurrence. Nevertheless, all children who have received GH, in particular cancer survivors and those receiving GH in adulthood, should be in surveillance programs to assess whether an increased rate od late-onset and rare tumours may occur.

Pathophysiology of radiation-induced growth hormone deficiency: efficacy and safety of GH replacement

Radiation-induced growth hormone deficiency (GHD) is primarily due to hypothalamic damage. GH secretion by the pituitary may be affected either secondary to some degree of quantitative deprivation of hypothalamic input or, if the radiation dose is high enough, by direct pituitary damage. As a consequence, the neurosecretory profile of GH secretion in an irradiated patient remains pulsatile and qualitatively intact. The frequency of pulse generation is unaffected, but the amplitude of the GH pulses is markedly reduced. Over the last 25 years, the final heights achieved by children receiving GH replacement for radiation-induced GHD have improved; these improvements are attributable to refinements in GH dosing schedules, increased use of GnRH analogues for radiation-induced precocious puberty, and a reduced time interval between completion of irradiation and initiation of GH therapy. When retested at the completion of growth, 80-90% of these teenagers are likely to prove severely GH deficient and, therefore, will potentially benefit from GH replacement in adult life. Such long-term GH treatment in patients treated previously for a brain tumor means that critical and continuous surveillance must be devoted to the risk of tumor recurrence and the possibility of second neoplasms.

Cancer Risk Following Growth Hormone Use in Childhood

The therapeutic use of growth hormone (GH) has caused concern, as it is anabolic and mitogenic, and its effector hormone, insulin-like growth factor (IGF)-I is anti-apoptotic. As both hormones can cause proliferation of normal and malignant cells, the possibility that GH therapy may induce cancer, increase the risk of tumour recurrence in those previously treated for a malignancy, or increase the risk of cancer in those with a predisposition, has resulted in concerns over its use. There are theoretical and epidemiological reasons that suggest GH and IGF-I may be important in tumour formation and proliferation. Malignant tumours have been induced in animals exposed to supraphysiological doses of GH, whereas hypophysectomy appears to protect animals from carcinogen-induced neoplasms. In vitro, proliferation and transformation of normal haemopoetic and leukaemic cells occurs with supraphysiological doses of GH, but not with physiological levels. IGF, IGF binding proteins (IGFBP) and IGFBP proteases influence the proliferation of cancer cells in vitro; however, GH is probably not involved in this process. Epidemiological studies have suggested an association between levels of IGF-I and cancer, and an inverse relationship between IGFBP-3 and cancer; however, these associations have been inconsistent. A number of studies have been undertaken to determine the risk of the development of cancer in children treated with GH, either de novo, or the recurrence of cancer in those previously treated for a malignancy. Despite early concerns following a report of a cluster of cases of leukaemia in recipients of GH, there appears to be no increased risk for the development of leukaemia in those treated with GH unless there is an underlying predisposition. Even in children with a primary diagnosis of cancer, subsequent GH use does not appear to increase the risk of tumour recurrence. However, a recent follow-up of pituitary GH recipients has suggested an increase in colorectal cancer. In addition, follow-up of oncology patients has suggested an increase in second neoplasms in those who also received GH therapy. These studies emphasise the importance of continued surveillance both internationally with established databases and also nationally through single-centre studies.

Disturbances in craniofacial morphology in children treated for solid tumors

Because of increased survival rates in childhood cancer, special interest has been focused on the side-effects of the therapy and the quality of life in long-term survivors. Our aim was to investigate craniofacial growth in children who had received different kinds of antineoplastic therapies for solid tumors. A total of 40 children treated in the Turku University Central Hospital were examined and divided into three different groups. Group 1 comprised 18 children treated for intracranial tumors with cranial irradiation (CRI) and chemotherapy (CT) including alkylating agents. Seven children out of 18 in this group received growth hormone (GH) therapy. In Group 2, 11 children with extracranial solid tumors also received multiagent CT including alkylating agents, but no CRI. Group 3 consisted of 11 children treated for Wilm's tumor with CT, which did not include alkylating agents or CRI. A total of 19 linear and four angular variables from the lateral cephalograms of the subjects were measured. Most deviations in craniofacial structures were found in children treated with combined CRI and multiagent CT. All disturbances were seen in the vertical measurements which were reduced when compared to the matched controls. It seems reasonable to assume that impaired growth following combined radio- and chemotherapy, as well as GH treatment, particularly affects cartilage-mediated growth. However, the deviations seen in the present study were fairly minor and did not usually require clinical consideration.

Increased incidence of neoplasia in patients with pituitary adenomas. The Pituitary Study Group

OBJECTIVE

The goal of our study was to determine the rate of neoplasms in patients with other pituitary adenomas (non-functioning and prolactinomas) in comparison with acromegaly which is known to favour the development of neoplasia.

DESIGN AND PATIENTS

We reviewed clinical records for 220 patients with acromegaly, 151 patients with non-functioning pituitary adenoma (NF) and 98 patients with prolactinomas. Incidence rates of cancer for patients with pituitary tumours were calculated per person-years of follow-up study. These rates were then compared with sex and age adjusted incidence rates reported by National Tumour Registry. An internal control group of 163 subjects with a non-neoplastic condition, i.e. Graves' disease followed chronically in the same clinic was also studied. The ratios observed to expected were expressed as standardized incidence rates (SIR). The only significant difference between the acromegalic and other pituitary tumours patients was in hypopituitarism, present in 18.2% (acromegaly) 47% (NF) and 18.6% (prolactinomas).

RESULTS

Twenty-three malignant tumours were registered in 19 acromegalics (1 Hodgkin disease, 1 myelogenous leukaemia, 1 lymphocytic leukaemia, 3 papillary thyroid carcinomas, 1 ovarian carcinoma, 2 colorectal carcinoma, 1 renal cell carcinoma, 4 cervical carcinoma, 2 skin cancers, 2 pancreatic carcinoma, 4 breast carcinoma, 1 bladder carcinoma). Three acromegalics harboured two malignancies. Patients with acromegaly had a 3.39-fold increased rate of malignant tumours compared with the general population and a 3.21-fold increased rate compared with our internal control group. Eleven malignant tumours were found in patients with NF-pituitary adenomas and 2 in prolactinoma patients (1 lymphoma, 1 multiple myeloma, 1 colonic cancer, 1 renal cell cancer, 1 stomach cancer, 2 lung cancers, 1 cervix carcinoma, 1 breast cancer, 1 testicular carcinoma and 3 melanoma). Patients with NF pituitary adenomas had a 3.91-fold increased rate of malignant tumours compared with the general population and 4.07-fold increase compared with the internal control group. Patients harbouring prolactinomas did not have an increased incidence rate of malignancy compared with the general population or our internal controls. Female patients with acromegaly and male patients with NF-pituitary adenoma had higher incidences of neoplasia.

CONCLUSION

We have demonstrated that the overall incidence of malignant tumours in patients with non-functioning pituitary adenomas and acromegaly is significantly higher than expected for general population and for our internal control group.