Background: The tyrosine kinase inhibitors (TKI) vandetanib and cabozantinib are approved as targeted therapies in advanced medullary thyroid carcinoma (MTC) with symptoms or high tumour burden. Only recently, toxicity in long-time TKI usage was analysed. However, little is known about the impact of TKI discontinuation on MTC disease course after longer-term therapy. Here, we report our experience in a series of 7 MTC patients with vandetanib treatment of up to 87 months followed by discontinuation for concerns of toxicity or due to side-effects. The discontinuation of TKI therapy is a relevant clinical scenario. To our knowledge we present the largest single center series on an important aspect of TKI management.

Methods: Retrospective analysis of MTC patients with continued discontinuation of vandetanib treatment in a tertiary referral endocrine tumour centre. Analysis included a review of patients’ records for TKI indication, and treatment response as well indications for continued TKI discontinuation and follow-up by clinical assessment, calcitonin and CEA doubling times as well as imaging (ultrasound, CT).

Results: Seven MTC patients [6 sporadic MTC, 1 Multiple Endocrine Neplasie Type 2a (MEN2a)] with previous vandetanib treatment (median: 41 months; range 7-87 months) and continued TKI discontinuation were identified out of 161 analysed MTC files. TKI treatment was initiated due to high tumour burden and symptoms or RECIST (Response Evaluation Criteria In Solid Tumors) progression in all patients. Two patients (29%) remained stable after discontinuation of vandetanib until now (follow-up of 47 and 61 months). Both patients had been on TKI therapy for 73 and 58 months. Five patients (71%) developed progressive disease after TKI discontinuation. In 2 patients, vandetanib was restarted after 45 and 52 months resulting again in disease control. One patient was enrolled in a new RET kinase inhibitor trial after 45 months of vandetanib discontinuation. Two patients declined restart of treatment due to mental health issues leading to discontinuation of vandetanib in the first place (after 7 and 38 months of treatment) and both patients died of rapidly progressive disease. At time points of tumour progression, calcitonin-doubling time (CDT) was < 2 years in all patients.

Conclusion: This case series suggests that discontinuation of long-term vandetanib treatment with documented stable disease does not automatically result in rapid disease progression but may be followed by prolonged “TKI free” stable disease in individual patients. Analysis of calcitonin and CDT during discontinuation is indicated as it will unmask tumour progression earlier than imaging. Restart with the same TKI is possible in case of progression.