Neoadjuvant Selective RET Inhibitor for Medullary Thyroid Cancer: A Case Series

Materials and Methods

This retrospective consecutive case series was conducted under institutional IRB protocol PA-14-1082, by which patient consent was waived. All patients with locoregionally advanced RET-mutated MTC who underwent neoadjuvant treatment with an RET-specific inhibitor before surgery were reviewed. The first patient was treated with the study drug LOXO-292 (provided by Loxo Oncology) after an investigational drug application approval by the FDA, which has been previously published. ³ The remaining three patients were treated outside of a clinical trial. The timing of medical and surgical intervention was determined by the treating provider based on tumor characteristics and response to treatment. Staging was based on the American Joint Committee on Cancer, eighth edition, and the International Union Against Cancer. ⁶

The primary outcomes were pre- versus postneoadjuvant response assessments using response evaluation criteria in solid tumors (RECIST) and a thyroid surgical morbidity score. RECIST was calculated by a board-certified neuroradiologist using the RECIST version 1.1. ⁷ RECIST was defined as the percentage decrease in longest diameter of locoregional target lesions from baseline to presurgery (after neoadjuvant treatment). To measure the potential clinical impact of neoadjuvant treatment response, change in thyroid neck surgical morbidity score was assessed using CT neck with contrast (1.25 mm cuts) at baseline and preoperatively at the completion of neoadjuvant therapy. The thyroid surgical morbidity scoring system was assessed on a scale of 0–4, and it was not validated (Supplementary Data). Secondary outcomes included locoregional disease-free survival extending from date of surgery and changes in calcitonin and carcinoembryonic antigen (CEA) levels. Adverse events were defined using the National Cancer Institute's Common Terminology Criteria for Adverse Events, version 5.0. ⁸

Results

We identified four patients who were treated with neoadjuvant selpercatinib followed by surgery between August 2018 and May 2022. Patients underwent 4–6 months of neoadjuvant treatment. Selpercatinib was typically held three days before surgery. Patients were followed for a median of 2.0 years from initiation of treatment (range 0.7–3.6).

Patient 1 has been previously reported, and his outcome has been updated for this series. ³ A 20-year-old man presented with stage IVC, cT4bN1bM1 sporadic MTC with somatic RET M918T mutation and RET deletion Y900_S904delisP. The patient's disease was deemed unresectable due to encasement of the subclavian artery, and his left recurrent laryngeal nerve (RLN) was at high risk, although his preoperative vocal cord function was intact. After 156 days of neoadjuvant selpercatinib (160 mg twice daily), he had a 55% RECIST and a reduction of his surgical morbidity score from 4 to 2.

He underwent total thyroidectomy, bilateral neck dissection, and median sternotomy with superior mediastinal dissection. An R1 (microscopic residual disease) gross total resection was achieved, and bilateral RLNs were preserved. Selpercatinib was electively restarted 9 days after surgery for control of pulmonary and spinal metastasis. The patient continues to receive selpercatinib at last follow-up at 3.6 years, with no evidence of structural locoregional disease. He experienced only grade 1 transaminitis, not requiring dose modification.

Patient 2, a 60-year-old man, presented with stage IVC, cT4aN1bM1 sporadic MTC. Somatic alterations were notable for the RET M918T mutation, loss of CDKN2C, and gain of CKS1B. He was treated with selpercatinib for 120 days, initially at 160 mg twice daily but reduced to 80 mg due to grade 2 lymphocytopenia and grade 1 xerostomia, leukopenia, and thrombocytopenia. He achieved a 32% RECIST and a reduction in surgical morbidity score from 3 to 2. He underwent total thyroidectomy and bilateral neck dissection.

Before neoadjuvant therapy, his bilateral RLNs were at high risk (extrathyroidal/extranodal disease in bilateral tracheoesophageal grooves) but he ultimately only required sacrifice of the right RLN. His right RLN was hypomobile preoperatively and expectedly immobile postoperatively, but his left cord was intact postoperatively. Treatment with selpercatinib was resumed at 80 mg twice a day 1.2 years after surgery due to progression of pulmonary disease. He continues to receive this treatment 2.2 years after surgery, with CT imaging suggesting no evidence of structural disease in the neck.

Patient 3 is a 55-year-old man with sporadic MTC metastatic to liver, lung, and bone who underwent total thyroidectomy and bilateral neck dissection at an outside hospital. He presented 1.5 years later with stage IVC, ypT1bN1bM1 disease with progression of cervical and mediastinal metastasis but intact vocal fold function. Somatic RET C630R mutation was noted. He was preoperatively treated with selpercatinib at 160 mg twice a day, but the drug was held for 2 days and restarted at 80 mg along with an antihypertensive medication due to transient grade 2 hypertension.

After 178 days of selpercatinib treatment, he had a 31% RECIST and underwent revision left neck dissection. Although his surgical morbidity score was unchanged at 2, he had a significant reduction in disease burden in the tracheoesophageal groove, with preservation of his RLN. Due to progressive liver and osseous metastasis, selpercatinib was restarted at 80 mg 0.6 years after surgery, which he has continued without further adverse events. There was no evidence of structural disease in the neck at last follow-up 1.4 years after surgery.

Patient 4, a 40-year-old woman, presented with stage IVA ypT1bN1bM0 sporadic MTC with somatic RET M918T mutation after initially undergoing total thyroidectomy and bilateral neck dissection 9 months earlier at an outside hospital. Her initial surgery was notable for significant residual disease invading the left RLN. She reported hoarseness that improved with neoadjuvant treatment, although bilateral vocal fold motion was grossly intact. After 176 days of selpercatinib at 160 mg twice a day, she had a 24% RECIST, considered stable disease.

She underwent revision of left central and lateral neck dissection with an R1 gross total resection and preservation of the RLN. Although her surgical morbidity score did not change from 2, the significantly at-risk RLN was preserved. At 0.7 years of follow-up, she has no structural evidence of disease in the neck.

A summary of the patient demographic and treatment outcomes is given in Table 1. The median time from initiation of neoadjuvant RET inhibitor treatment to surgery was 168 days (range 120–178). Pre- and postneoadjuvant treatment (preoperative) CT scans with contrast for each patient are shown in Figure 1. There was a median 32% RECIST (range 24–55%), with 3 of 4 patients achieving a partial response. The median surgical morbidity score improvement was 1 (range 0–2) after neoadjuvant therapy. There were an estimated 5 RLNs at risk before neoadjuvant therapy, but only 1 RLN was ultimately sacrificed intraoperatively.

OPEN IN VIEWER

FIG. 1.

Neck CT scans with contrast before and after neoadjuvant therapy with a selective RET inhibitor.

The trends in calcitonin and CEA levels throughout treatment are shown in Supplementary Figure S1. Median calcitonin and CEA values at last follow-up were 4.4% (range 2.6–60%) and 56% (range 7.6–196%) of the preneoadjuvant treatment values, respectively. Treatment was transiently held in one patient for hypertension. Otherwise, there were no perioperative complications. All patients are alive at a median follow-up of 2.0 years (range 0.7–3.6) without structural locoregional disease.

Discussion

This report represents the first case series of patients with locoregionally advanced MTC treated with neoadjuvant selective RET inhibitor. Only a single case report using this approach has been published to date—a patient included in this series with longer follow-up. ³ In this study, significant radiographic responses and surgical preservation of major neurovascular structures were observed after presurgical selpercatinib therapy.

The pathology was marked by fibroadipose deposits, fibrosis, perineural invasion, and/or extranodal extension, resulting in technically challenging surgery. Nevertheless, four RLNs were preserved despite being significantly at risk before initiation of RET inhibitor. Of four patients, three restarted selpercatinib therapy after surgery due to progression of distant disease. Adverse events were limited to 2 patients with grade 1 and 2 patients with grade 2. Locoregional disease control was achieved at last follow-up.

This study is limited to four patients with somatic RET-mutated MTC. While RECIST was assessed, no validated methodology currently exists for assessing changes in surgical morbidity with neoadjuvant treatment. Adverse effects were minimal, but the full range of treatment sequala was likely not represented in this small sample. While neoadjuvant targeted treatments have shown promise with other thyroid cancers, this series does not compare different treatments and the application of this approach must be determined on an individual basis. ⁹ A clinical trial is currently underway to further study neoadjuvant selpercatinib treatment for RET-altered MTC (NCT04759911). ¹⁰

Conclusions

Patients with locoregionally advanced RET-altered MTC may achieve locoregional disease control with reduction in surgical morbidity using a selective RET inhibitor in the neoadjuvant setting. A clinical trial is ongoing to further evaluate this novel approach.