1761
Correlation Between Somatostatin Treatment and Elevated Hepatic Fat Fraction on MRI in Patients with Neuroendocrine Tumors1Radiology, Stanford University, Stanford, CA, United States, 2Oncology, Stanford University, Stanford, CA, United States
Synopsis
Up to 90% of neuroendocrine tumor patients have metastatic disease in the liver at diagnosis. These patients are treated with somatostatin analog therapy and monitored with CT or MRI. We demonstrate in a retrospective study that somatostatin therapy is associated with the development of elevated liver fat fraction on MRI. Furthermore, preliminary results suggest that hepatic steatosis decreases lesion detectability on CT compared to MRI. Studies are ongoing to determine the severity of steatosis and relationship to cumulative somatostatin dose, variation in fat fraction over time and the response to change in medication, and effect on liver function.
Purpose
Although neuroendocrine tumors are rare, they are often diagnosed late due to the nonspecific symptoms. Thus, up to 90% patients have metastatic disease in the liver at diagnosis (1). One of the cornerstones of therapy for these patients are somatostatin analogs, which control hormone-associated symptoms like flushing and diarrhea, and growth and disease progression (2). Disease burden is monitored with imaging and, with symptoms determines treatment options and prognosis. At our institution, there have been several cases of neuroendocrine patients developing hepatic steatosis while on somatostatin therapy, which is currently not a known side effect of these drugs. We hypothesized that somatostatin analogs are associated with the development of elevated liver fat fraction on MRI and that this reduces detectability of liver metastases on CT relative to MRI.Methods
This was a retrospective evaluation of a cohort of adult neuroendocrine tumor patients with baseline MRI, non-contrast CT, or ultrasound of the liver prior to initiating somatostatin analog therapy and follow up MRI exams for at least one year after starting therapy. Presence or absence of hepatic steatosis was determined on baseline imaging (patients were positive for steatosis if fat fraction was greater than 5% on MRI, the liver attenuation on CT was less than 40 HU or at least 10 HU less than the spleen, or increased echogenicity of the liver relative to the kidney on ultrasound). Fat fractions were recorded for follow up MRI exams based on in-phase and out-of-phase imaging. A subset of 4 patients with steatosis and MRI and CT performed within 3 months of each other was analyzed for number of metastases detected by CT versus number detected by MRI.Results
Of 38 patients with baseline imaging and at least one year of follow up MRIs, 30 were not steatotic at baseline and 7 (19%) did have steatosis (consistent with estimated steatosis rates in the general population3,4). One patient had waxing and waning mild steatosis and was excluded. 15 of the 30 patients who did not have hepatic steatosis at baseline developed it on MRI within one year of starting somatostatin therapy. Of the 7 patients with steatosis at baseline, 6 continued to have steatosis and 1 improved to normal level. Analysis using McNemar’s test show statistically significant association between somatostatin analog therapy and development of hepatic steatosis (Table 1, Figure 1). Further analysis of the subset of 27 patients who had baseline MRI rather than CT or ultrasound demonstrated 37% (19-58%) one year incidence rate of steatosis, with older patients who have borderline fat fractions at baseline more likely to develop steatosis (Figure 2). Finally, in our analysis of 4 steatotic patients for lesion detectability, we found that at best, CT was able to detect up to 30% of the lesions seen on MRI (Table 2).Discussion
Our results show a statistically significant association between somatostatin therapy and hepatic steatosis with annual incidence rate of steatosis higher than for the general population, which ranges from 2-10% (5-7). Additionally, of 29 patients who were initially excluded due to insufficient MRI follow up, 10 had steatosis on MRI performed greater than one year from the start of somatostatin therapy or had evidence of steatosis on CT, suggesting that the actual rate of steatosis may be even higher. Although hepatic steatosis has been reported as an adverse effect of an overdose of somatostatin analog (8), to our knowledge, this is the first study that demonstrates an association between somatostatin analog therapy and detectable hepatic steatosis on MRI. Studies are ongoing to determine the severity of steatosis and relationship to cumulative somatostatin dose, variation in fat fraction over time and the response to change in medication, and effect on liver function. A small subset of neuroendocrine patients with steatosis also had decreased detectability of liver metastases on CT compared to MRI (Figure 3), suggesting that CT may not be sufficient to monitor these patients. We are further evaluating this in a larger patient cohort. There are several limitations to this work, including relatively small sample size, inconsistent frequency and modality of imaging baselines and follow ups, and MRI scanner and protocol variability. This study forms the basis for subsequent prospective studies at a larger scale to track the development of hepatic steatosis over time in patients on somatostatin analog therapy and the effects on liver function and disease detection.Conclusion
Somatostatin analog therapy, which is commonly used in neuroendocrine tumor patients to control disease symptoms and progress, is associated with the development of hepatic steatosis, which can be detected and monitored by MRI. Preliminary data also suggests that presence of steatosis decreases metastasis detectability on CT relative to MRI.Acknowledgements
No acknowledgement found.References
1. Colquhoun S. Neuroendocrine tumors with hepatic metastases: A review of evolving treatment options. Liver Research 2018;2(2):92-99.
2. Harring TR, Nguyen NT, Goss JA, O'Mahony CA. Treatment of liver metastases in patients with neuroendocrine tumors: a comprehensive review. Int J Hepatol. 2011;2011:154541.
3. Angulo P. Nonalcoholic fatty liver disease. N Engl J Med. 2002;346(16):1221-1231.
4. Fraum TJ, Ludwig DR, Kilian S, et al. Epidemiology of Hepatic Steatosis at a Tertiary Care Center: An MRI-based Analysis. Acad Radiol. 2018;25(3):317-327.
5. Kanwal F, Kramer JR, Duan Z, Yu X, White D, El-Serag HB. Trends in the Burden of Nonalcoholic Fatty Liver Disease in a United States Cohort of Veterans. Clin Gastroenterol Hepatol. 2016;14(2):301-308.e301-302.
6. Perumpail BJ, Khan MA, Yoo ER, Cholankeril G, Kim D, Ahmed A. Clinical epidemiology and disease burden of nonalcoholic fatty liver disease. World J Gastroenterol. 2017;23(47):8263-8276.
7. Younossi ZM, Koenig AB, Abdelatif D, Fazel Y, Henry L, Wymer M. Global epidemiology of nonalcoholic fatty liver disease-Meta-analytic assessment of prevalence, incidence, and outcomes. Hepatology. 2016;64(1):73-84.
8. Sandostatin. https://www.accessdata.fda.gov/drugsatfda_docs/label/2012/019667s061lbl.pdf. Accessed 11/6/2018, 2018.
Figures
