Colorectal neoplasia in acromegaly

Diabetes and Colorectal Cancer Risk: Clinical and Therapeutic Implications

Several epidemiological studies have identified diabetes as a risk factor for colorectal cancer (CRC). The potential pathophysiological mechanisms of this association include hyperinsulinemia, insulin-like growth factor (IGF) axis, hyperglycemia, inflammation induced by adipose tissue dysfunction, gastrointestinal motility disorder, and impaired immunological surveillance. Several studies have shown that underlying diabetes adversely affects the prognosis of patients with CRC. This review explores the novel anticancer agents targeting IGF-1R and receptor for advanced glycation end products (RAGE), both of which play a vital role in diabetes-induced colorectal tumorigenesis. Inhibitors of IGF-1R and RAGE are expected to become promising therapeutic choices, particularly for CRC patients with diabetes. Furthermore, hypoglycemic therapy is associated with the incidence of CRC. Selection of appropriate hypoglycemic agents, which can reduce the risk of CRC in diabetic patients, is an unmet issue. Therefore, this review mainly summarizes the current studies concerning the connections among diabetes, hypoglycemic therapy, and CRC as well as provides a synthesis of the underlying pathophysiological mechanisms. Our synthesis provides a theoretical basis for rational use of hypoglycemic therapies and early diagnosis and treatment of diabetes-related CRC.

Progress in understanding of relationship between diabetes and colorectal cancer

Several epidemiological studies have suggested that diabetes is closely associated with an increased risk of colorectal cancer and diabetes could be regarded as an independent risk factor for colorectal cancer. Potential pathophysiological mechanisms connecting diabetes and colorectal cancer include hyperglycemia, hyperinsulinemia, and insulin-like growth factor axis, chronic inflammation and oxidative stress, gastrointestinal motility disorder, and impaired immunological surveillance. Meanwhile, multiple studies have revealed that diabetes is negatively related to the prognosis of patients with colorectal cancer. This review mainly summarizes the current studies concerning the linkages between diabetes and colorectal cancer and the underlying pathophysiological mechanisms, so as to provide a theoretical basis for rational use of antidiabetic drugs and early diagnosis of diabetes-related colorectal cancer.

Adiponectin, Leptin, IGF-1, and Tumor Necrosis Factor Alpha As Potential Serum Biomarkers for Non-Invasive Diagnosis of Colorectal Adenoma in African Americans

The potential role of adiponectin, leptin, IGF-1, and tumor necrosis factor alpha (TNF-α) as biomarkers in colorectal adenoma is not clear. Therefore, we aimed to investigate the blood serum levels of these biomarkers in colorectal adenoma. The case-control study consisted of serum from 180 African American patients with colon adenoma (cases) and 198 healthy African Americans (controls) at Howard University Hospital. We used ELISA for adiponectin, leptin, IGF-1, and TNF-α detection and quantification. Statistical analysis was performed by t-test and multivariate logistic regression. The respective differences in median leptin, adiponectin, IGF-1, and TNF-α levels between control and case groups (13.9 vs. 16.4), (11.3 vs. 46.0), (4.5 vs. 12.9), and (71.4 vs. 130.8) were statistically significant (P < 0.05). In a multivariate model, the odds ratio for adiponectin, TNF-α, and IGF-1 were 2.0 (95% CI = 1.6-2.5; P < 0.001), 1.5 (95% CI = 1.5(1.1-2.0); P = 0.004), and 1.6 (95% CI = 1.3-2.0; P < 0.001), respectively. There was a positive correlation between serum adiponectin and IGF-1 concentrations with age (r = 0.17, P < 0.001 and r = 0.13, P = 0.009), TNF-α, IGF-1, and leptin concentration with body mass index (BMI) (r = 0.44, P < 0.001; r = 0.11, P = 0.03; and r = 0.48, P < 0.001), respectively. Also, there was a negative correlation between adiponectin and leptin concentrations with BMI (r = -0.40, P < 0.001), respectively. These data support the hypothesis that adiponectin, IGF-1, and TNF-α high levels correlate with higher risk of colon adenoma and can thus be used for colorectal adenomas risk assessment.

Growth hormone and risk for cardiac tumors in Carney complex

Carney complex (CNC) is a multiple neoplasia syndrome that is caused mostly by PRKAR1A mutations. Cardiac myxomas are the leading cause of mortality in CNC patients who, in addition, often develop growth hormone (GH) excess. We studied patients with CNC, who were observed for over a period of 20 years (1995-2015) for the development of both GH excess and cardiac myxomas. GH secretion was evaluated by standard testing; dedicated cardiovascular imaging was used to detect cardiac abnormalities. Four excised cardiac myxomas were tested for the expression of insulin-like growth factor-1 (IGF-1). A total of 99 CNC patients (97 with a PRKAR1A mutation) were included in the study with a mean age of 25.8 ± 16.6 years at presentation. Over an observed mean follow-up of 25.8 years, 60% of patients with GH excess (n = 46) developed a cardiac myxoma compared with only 36% of those without GH excess (n = 54) (P = 0.016). Overall, patients with GH excess were also more likely to have a tumor vs those with normal GH secretion (OR: 2.78, 95% CI: 1.23-6.29; P = 0.014). IGF-1 mRNA and protein were higher in CNC myxomas than in normal heart tissue. We conclude that the development of cardiac myxomas in CNC may be associated with increased GH secretion, in a manner analogous to the association between fibrous dysplasia and GH excess in McCune-Albright syndrome, a condition similar to CNC. We speculate that treatment of GH excess in patients with CNC may reduce the likelihood of cardiac myxoma formation and/or recurrence of this tumor.

Growth hormone and cancer: GH production and action in glioma?

The hypersecretion of pituitary growth hormone (GH) is associated with an increased risk of cancer, while reducing pituitary GH signaling reduces this risk. Roles for pituitary GH in cancer are therefore well established. The expression of the GH gene is, however, not confined to the pituitary gland and it is now known to occur in many extrapituitary tissues, in which it has local autocrine or paracrine actions, rather than endocrine function. It is, for instance, expressed in cancers of the prostate, lung, skin, endometrium and colon. The oncogenicity of autocrine GH may also be greater than that induced by endocrine or exogenous GH, as higher concentrations of GHR antagonists are required to inhibit its actions. This may reflect the fact that autocrine GH is thought to act at intracellular receptors directly after synthesis, in compartments not readily accessible to endocrine (or exogenous) GH. The roles and actions of extrapituitary GH in cancer may therefore differ from those of pituitary GH. The possibility that GH may be expressed and act in glioma tumors was therefore examined by immunohistochemistry. These results demonstrate, for the first time, the presence of abundant GH- and GH receptor (GHR-) immunoreactivity in glioma, in which they were co-localized in cytoplasmic but not nuclear compartments. These results demonstrate that glioma differs from most cancers in lacking nuclear GHRs, but GH is nevertheless likely to have autocrine or paracrine actions in the induction and progression of glioma.

Acromegaly, growth hormone and cancer risk

Acromegaly is an endocrine disorder characterized by sustained hypersecretion of growth hormone (GH) with concomitant elevation of insulin-like growth factor I (IGF-I) associated with premature mortality from cardiopulmonary diseases and certain malignancies. In particular, there is a two-fold increased risk of developing colorectal cancer. Possible mechanisms underlying this association include elevated levels of circulating GH and IGF-I, but several other plausible processes may be relevant. In a parallel literature, there has been debate whether GH replacement therapy is associated with increased cancer risk in three scenarios: (1) tumour recurrence in children with previously treated cancer; (2) second neoplasms (SNs) in survivors of childhood cancer treated with GH; and (3) de-novo cancer in non-cancer patients treated with GH. The general evidence suggests no increased risk in scenario 1. Through a maze of complex study designs, there is inconclusive evidence of a very modest increase in cancer risk in treated GH-deficiency patients in scenarios 2 and 3, but it is likely that the cumulative risk equates to that of the general population. This emphasizes the need for patient selection balanced against the known morbidity of untreated GH deficiency.

The insulin-like growth factor system and colorectal cancer: clinical and experimental evidence

OBJECTIVE

The aim of this review is to clarify the involvement of the insulin-like growth factor (IGF) system in the development of colorectal malignancy.

MATERIALS AND METHODS

Medline searches were used to identify key articles relating the IGF system with the development of colorectal cancer.

RESULTS

The IGF system has been linked to colorectal malignancy by a convergence of data from epidemiological, clinical and laboratory-based sources.

CONCLUSION

Further work is needed to characterise the IGF system expression in the colon. Such clarification could lead to the identification of targets that can be manipulated for clinical advantage.

Hodgkin lymphoma in temporal association with growth hormone replacement

The association between growth hormone (GH) replacement and malignancy has long been debated. We report a case of Hodgkin lymphoma that developed in temporal association with the initiation of GH replacement in a 57-year-old woman with panhypopituitarism secondary to a non-secretory pituitary macroadenoma. Treatment of her pituitary tumor included transphenoidal surgery, external beam radiation, Bromocriptine and Cabergaline therapy. In addition to replacement steroid, thyroid and sex hormones, she insisted on GH replacement. Approximately 2 years after GH initiation, the diagnosis of Hodgkin lymphoma was made. Although the exact contribution of GH to the development of Hodgkin disease in our patient is unclear and a causal effect cannot be concluded, the temporal association is suggestive, and warrants reporting as part of ongoing surveillance for potential complications of GH replacement.

Acromegaly and cancer

In recent years, it has become increasingly recognized that acromegaly is associated with an increased prevalence of colorectal cancer and pre-malignant tubular adenomas. The aetiology of these tumours is unknown but is likely to reflect increased levels of both insulin-like growth factor I (IGF-I), which is implicated in the development of sporadic colorectal cancer, and environmental factors, such as the bile acid deoxycholic acid. There is also evidence to suggest that the prevalence of breast and perhaps haematological malignancies might be increased in acromegaly, although these associations have been based on mostly small epidemiological surveys and clarification will come in the future once large-scale epidemiological studies have been completed.

Risk of colorectal neoplasm in patients with acromegaly and its relationship with serum growth hormone levels

OBJECTIVES

Acromegalics have been reported to be at an increased risk of colorectal neoplasm. However, the magnitude of the risk is still controversial and the mechanism has not been fully investigated. In this study, we attempted to determine the magnitude of the association between acromegaly and colorectal lesions after taking into account age, gender, smoking status, and treatment status. In addition, we assessed the relationship between colorectal lesions and serum growth hormone (GH) levels in acromegalics.

METHODS

We conducted a case-control study by using 19 consecutive untreated patients (male:female = 11:8) who were newly diagnosed with acromegaly between 1990 and 2000. All patients underwent colonoscopy and received a histological diagnosis of colorectal lesions. Prevalence of hyperplastic polyp, adenoma, and carcinoma were compared with the prevalence in 76 controls matched for gender, age, and smoking status. Serum GH levels were compared between acromegalic patients with and without each type of colorectal lesion.

RESULTS

The prevalence of hyperplastic polyp, adenoma, and carcinoma were significantly higher in the acromegalic patients compared to the controls (p < 0.05, odds ratios; 8.3, 4.2, and 9.8, respectively). In acromegalics, the presence of hyperplastic polyps and carcinomas were significantly associated with higher serum GH levels after adjusting for the other lesions and age (p < 0.05).

CONCLUSIONS

After controlling for age, gender, smoking status, and treatment status, acromegaly was associated with significantly higher prevalence of colorectal hyperplastic polyp, adenoma, and carcinoma. High serum GH levels may be associated with the presence of hyperplastic polyp and carcinoma.

Determination of large bowel length and loop complexity in patients with acromegaly undergoing screening colonoscopy

BACKGROUND

Patients with acromegaly are at moderately increased risk of developing colorectal cancer and may be considered for screening colonoscopy. In turn, large bowel dimensions may be increased in these patients, factors that predict for increased risk of serious complications such as perforation.

OBJECTIVE

To evaluate this risk potential, we measured large bowel length and loop complexity using magnetic endoscopic imaging (MEI).

DESIGN

Case-control study in 25 unselected patients with acromegaly (mean age 56 years) vs. 41 nonacromegalic controls (mean age 60 years) undergoing screening colonoscopy.

MEASUREMENTS

MEI parameters were determined and age- and sex-adjusted mean differences calculated. The dependency of total large bowel length on various demographic and disease-related factors (e.g. GH exposure, IGF-I and IGFBP-3 concentrations) was assessed using regression techniques.

RESULTS

Total large bowel length was increased by 20%[95% confidence interval (CI) 9-31%] in patients with acromegaly compared with controls (unadjusted and adjusted; P-values < 0.001). Acromegaly was also associated with increased time taken to reach the caecum (P = 0.01) and increased pelvic loop complexities (5/25 vs. 1/41, Fisher's exact test: P = 0.03). Total large bowel length was predicted by age at colonoscopy (P = 0.003) and patient height (P = 0.03), but not by surrogate biochemical markers of disease activity.

CONCLUSIONS

Acromegaly is associated with increased large bowel length and loop complexity making colonoscopy technically challenging, and theoretically increasing the risk of serious complications. Patients need to be counselled accordingly, and appropriate resources with experienced staff allocated.

Prevalence of colonic polyps is not increased in patients with acromegaly: analysis of 60 patients from India

BACKGROUND

There are conflicting data on the prevalence of colorectal adenomas in patients with acromegaly. It has been suggested that the risk of colorectal adenomas may be dependent on the geographic and ethnic origin of the patients.

MATERIALS

Sixty consecutive patients with active acromegaly due to somatotropinoma underwent colonoscopy prior to definitive surgery. They included 35 men and 25 women with a mean (+/- SD) age of 37.4 +/- 13.2 years and a lag time between symptomatology and diagnosis of acromegaly of 64.6 +/- 58.0 months. The control group included 160 patients (88 men, 72 women; mean age, 38.2 +/- 14.0 years) with a diagnosis of irritable bowel syndrome who were also subjected to colonoscopy. None of these patients had previous or family history of colonic neoplasm, bleeding or colonic surgery.

RESULTS

Colonoscopic examination was complete to the cecum in 52 patients (88%), and to the splenic flexure in the remaining 12%. In the control group, it was complete to the cecum in 144 patients (90%). Four of the acromegalic patients (6.7%) and five in the control group (3.1%; P=0.24) had hyperplastic polyps. No patient in either of the groups had adenomatous polyps or colonic adenocarcinoma. The group of acromegalic patients with and without polyps did not differ significantly in age, duration of disease, growth hormone levels or glycemic status. The number of skin tags, however, was significantly higher (P=0.04) in the polyp group as compared to those without polyps.

CONCLUSION

The present study has failed to demonstrate the high prevalence rate of colonic neoplasia in patients with acromegaly as compared to reports from the Western world.

Acromegaly and the risk of cancer

Recent studies suggest that acromegaly might predispose to an increased risk of benign and malignant neoplasms, thus influencing the final outcome of the disease. The exact mechanism of neoplastic events in acromegaly has not been completely clarified. Several studies indicate an autocrine-paracrine role for growth hormone (GH) and insulin-like growth factor-I (IGF-I) in the proliferation of normal and neoplastic cells. The paper reviews the results of molecular, clinical and epidemiological data supporting a role for GH-IGF-I action in colon, prostate, breast and lung carcinogenesis inpatients with acromegaly.

Colonic screening and surveillance

Screening for colorectal cancer has not obtained worldwide acceptance in spite of its proven survival benefit for average-risk persons and some high-risk groups. The incidence of and mortality from colorectal cancer are worrying in Europe as well as in the USA, Australia and Japan. The best evidence-based studies are those published on screening using faecal occult blood tests, endoscopic methods and different tumour markers having been evaluated to a lesser degree. Feasibility studies are necessary before massive screening can be undertaken because the results obtained from randomized studies may not be reproduced to a satisfactory degree in average- as well as high-risk populations. Primary prevention by dietary intervention and drugs has been studied in great detail, so far without any major breakthrough. This chapter will address different screening methods in populations with a varying risk of colorectal cancer, together with providing a short review of prevention and intervention strategies.

The prevalence and characteristics of colorectal neoplasia in acromegaly

An increased prevalence of colorectal neoplasia has been reported in acromegalic patients, and recommendations have been made for early colonoscopic screening and regular surveillance. This assumption, however, is frequently drawn from studies using selected control populations. To clarify colonoscopic management in these patients, we undertook a 2-center prospective screening colonoscopy study in 122 acromegalics (age range, 25-82 yr). In the absence of ideal age-matched controls, we calculated prevalence rates of occult adenocarcinomas and adenomas in the general population using cumulative data in the published literature from 8 autopsy studies (model 1, n = 3,559) and 4 screening colonoscopy studies (model 2, n= 810), applying linear regression models. Of the 115 patients with complete examinations, adenocarcinomas were discovered in 3 (2.6%), and at least 1 adenoma was found in 11, giving an overall prevalence of neoplasia of 12% (14 of 115). Prevalence rates for age bands 30-40, 40-49, 50-59, 60-69, and 70+ yr were 0%, 8%, 12%, 20%, and 21%, respectively. Compared with the 2 control models, the prevalence of occult colorectal cancer was not significantly increased (acromegalics vs. models 1 and 2, 2.6% vs. 2.3% and 0.9%), nor was there an increase in the prevalence of adenomas in any age band. Pathological characteristics showed some differences, in that adenomas in acromegalics tended to be right sided (68% vs. 57% and 56%), larger (for > or =10 mm, 27% vs. 13% and 9%), and of advanced histology (for tubulovillous, 27% vs. 4% and 22%). No associations were found between the presence of colonic neoplasia and the duration of disease, total GH exposure, cure status, and serum insulin-like growth factor I. This study has failed to demonstrate an increased prevalence of neoplasia in acromegalic patients compared with the expected prevalence in the general population and questions the need for an aggressive colonoscopic screening policy.

Circulating insulin-like growth factor II and colorectal adenomas

Circulating insulin-like growth factor I (IGF-I) and IGF-binding protein-3 (IGFBP-3) may be risk factors for the development of colorectal cancer. On the other hand, IGF-II and IGFBP-2 are overexpressed in colorectal carcinomas. These contrasting backgrounds led us to investigate the relationship between serum IGF-I, IGF-II, IGFBP-2, and IGFBP-3 and the presence of colorectal adenomas, known precursors of colorectal carcinoma, in 345 volunteers attending a screening flexible sigmoidoscopy trial (entry criteria: healthy, aged 55-64 yr). The most striking finding was an elevated mean serum IGF-II in individuals with adenomas (n = 52) compared with controls (mean difference, 139 ng/mL; 95% confidence intervals, 82, 196; P < 0.0001). Logistic regression adjusting for confounding factors confirmed the significant association between IGF-II and adenoma occurrence (P < 0.0001) and revealed an additional positive association with serum IGFBP-2 (P < 0.0001). However, there was no association found between either serum IGF-I and/or IGFBP-3 and the presence of adenomas. Additionally, in 31 individuals with adenomas in whom levels were determined pre- and postpolypectomy, there was a significant fall in mean IGF-II (P < 0.001) and IGFBP-2 (P < 0.001) after adenoma removal, but no difference in IGF-II and IGFBP-2 concentrations between repeated samples in 20 individuals without adenomas. Immunohistochemical studies demonstrated IGF-II expression in 83% of all adenomas, which contrasted with absent expression in normal colonic expression and hyperplastic polyps. This study has shown for the first time that serum IGF-II may be a tumor marker in individuals with colorectal adenomas. Further studies are needed to validate these relationships in larger populations, including individuals undergoing colonoscopy.

Pituitary incidentalomas

The optimal strategy for hormonal screening of a patient with any incidentally discovered pituitary mass is unknown. The authors' review of the endocrinologic literature supports the view that such patients are at slightly increased risk for morbidity and mortality. This risk implies a benefit of early diagnosis for at least for some of the disorders, suggesting the importance of case finding. Nevertheless, the data in Table 1 illustrate that clinically diagnosed hormone-secreting pituitary tumors are far less common than incidentalomas. Clinically, one cannot accurately determine the approximately 0.5% of patients with incidentaloma who are at increased risk among the vast majority who are not. Given the limitations of diagnostic tests, effective hormonal screening requires a sufficiently high pretest probability to limit the number of false-positive results. This condition is met to varying degrees in the patient with a small incidentally discovered pituitary mass but no signs or symptoms of hormone excess. Even the more common lesions, such as prolactinoma, are relatively rare. [table: see text] Subjecting patients to unnecessary testing and treatment is associated with risk. In addition to its initial cost, testing may result in further expense and harm as false-positive results are pursued, producing the "cascade effect" described by Mold and Stein as a "chain of events (which) tends to proceed with increasing momentum, so that the further it progresses the more difficult it is to stop." The extensive evaluations performed for some patients with incidentally discovered masses may reflect the unwillingness of many physicians to accept uncertainty, even in the case of an extremely unlikely diagnosis. This unwillingness may be driven, in part, by fear of potential malpractice liability, the failure to appreciate the influence of prevalence data on the interpretation of diagnostic testing, or other factors. The major justification for further evaluation of these patients is not so much to avoid morbidity and mortality for the rare patient who truly is at increased risk but to reassure patients in whom further testing is negative and the physician. Physicians must take care not to create inappropriate anxiety in patients by overemphasizing the importance of an incidental finding unless it is associated with a realistic clinical risk. The authors' recommendations are based on currently available information to minimize the untoward effects of the cascade. As evidence accumulates, these recommendations may need to be revised. The benefit of the diagnosis of an adrenal or pituitary disorder must be considered in the context of the patient's overall condition. Additional studies are needed to analyze the clinical utility of hormonal screening for these common radiologic findings. Data from these studies can be used to identify critical gaps in knowledge and to adopt the epidemiologic methods of evaluation of evidence that have been applied to preventive measures. One must be careful to recognize lead-time bias, in which survival can appear to be lengthened when screening simply advances the time of diagnosis, lengthening the period of time between diagnosis and death without any true prolongation of life; and length bias, which refers to the tendency of screening to detect a disproportionate number of cases of slowly progressive disease and to miss aggressive cases that, by virtue of rapid progression, are present in the population only briefly. Physicians must avoid the pitfalls of overestimation of disease prevalence and of the benefits of therapy resulting from advances in diagnostic imaging. Clinical judgment based on the best available evidence should be complemented and not replaced by laboratory data.