Blocking the Thyrotropin Receptor with K1-70 in a Patient with Follicular Thyroid Cancer,Graves' Disease,and Graves' Ophthalmopathy

Study patient

A 51-year-old female presented at the Mayo Clinic in October 2014 with progressive weight loss, fatigue, weakness, and a large left thyroid mass with associated left neck adenopathy. She smoked and had GO, significant thyrotoxicosis, and high levels of autoantibodies to the TSHR with high thyroid stimulating antibody (TSAb) activity. Imaging demonstrated a large suspicious 8 cm left thyroid lobe mass with left lateral neck adenopathy (about 3 cm). Following near-total thyroidectomy, pathology confirmed an 8.7 cm widely invasive FTC with soft tissue extension and associated multiple metastatic cervical lymph nodes. A thyrogen-stimulated ¹²³I whole-body scan demonstrated iodine avid metastatic foci within the neck, mediastinum, chest, and skeletal system. She received 250 mCi ¹³¹I therapy followed by thyrotropin (TSH) suppression with thyroxine (T4). The patient remained on T4 throughout the course of her disease with suppressed TSH levels and stable T4 and triiodothyronine (T3) levels. There were no significant elevations in T4 and T3 levels throughout the observation period. As T4 was not withdrawn, endogenous T4 and T3 levels could not be assessed. Tumor progression necessitated two further neck operations, debulking thoracotomies for mediastinal disease and a second dose of ¹³¹I. Her GO, in the context of ongoing smoking, high TSAb activity, selenium therapy, and multiple doses of ¹³¹I, remained active and progressive, with a clinical activity score (CAS) of 6/7 and diplopia, the latter precluding her ability to drive.

The patient's GO and rapidly progressive FTC with local recurrences were postulated to be influenced, at least in part, by her high TSAb activity. Attempts at mitigating the effects of TSAb on her disease with intravenous-immunoglobulin (IV-IG) were unsuccessful. Following IV–IG therapy in December 2016, the CAS was 6/7 with proptosis 21/21 compared with CAS 2/7 and proptosis 20/23 in October 2016. Furthermore, high-dose steroids administered in May 2018 at the time of further ¹³¹I therapy showed no clear effects. Genomic evaluations of her nodal disease identified NRAS (Q61R), RBM10 (E770fs*17), and TERT promoter (−124C>T) alterations, which, together with high TSAb activity, were likely associated causally with her advanced disease. Lenvatinib antiangiogenic therapy was initiated in January 2017, ∼2 years after her initial FTC diagnosis, with a partial tumor response, but she had dose-limiting toxicities (fatigue, chest pain, and decreased appetite). A multispecialty Tumor Board review recommended targeting the TSHR with the novel human monoclonal antibody, K1-70, a recombinant immunoglobulin G that is a specific and powerful TSHR antagonist designated for intramuscular (i.m.) or intravenous (i.v.) administration (3,4). A single patient expanded access application was granted by the U.S. Food and Drug Administration for this patient for therapy with K1-70, in combination with systemic antineoplastic therapies. The expanded access use was approved by the local Institutional Review Board.

K1-70 administration

In April 2017, the patient received her first dose of K1-70 (17.6 mg i.m.; Table 1). K1-70 was then administered at 3 weekly intervals with the dose increased to 40 mg i.m. and then to 120 mg i.m. and this blocked her high TSAb activity (Table 1; Fig. 1). Thereafter, doses of K1-70 were adjusted according to measurements of serum TSAb activity with the dose reduced to 60 mg i.m. when serum TSAb became undetectable, and increased to 120 mg i.m. when TSAb activity started to increase (Table 1).

OPEN IN VIEWER

FIG. 1.

(A) Effect of K1-70™ i.m. on TSAb and TSBAb activities in the patient with metastatic thyroid follicular cancer, Graves' disease, and GO. Reproduced with permission of the copyright holder, AV7 Ltd. Patient serum K1-70 level (mg/L) is shown with open triangles (Δ), patient serum TSAb activity (% stimulation) is shown with open circles (○), patient serum TSBAb activity (% inhibition) is shown with open squares (□), and TSAb activity was measured in an RSR™ bioassay using Chinese hamster ovary cells stably expressing human TSHR (5). Stimulation >180% relative to the response obtained with a healthy blood donor serum pool was considered positive. TSBAb activity was measured in an RSR TSBAb bioassay. The TSAb and TSBAb activities reflected “net” serum activities of the mixture of endogenous TSAb and administered K1-70. (B) The effect of blocking stimulation of the TSHR with the human monoclonal antibody K1-70 on moderate-to-severe GO. Baseline image from: (a) anterior, (b) oblique, and (c) lateral views. Postsurgery images from (d) anterior, (e) oblique, and (f) lateral views. See text and Table 1 for details. The image is reproduced with the patient's informed consent. GO, Graves' ophthalmopathy; i.m., intramuscular; TSAb, thyroid stimulating antibody; TSBAb, thyroid stimulating blocking activity; TSHR, thyrotropin receptor.

Laboratory measurements

Serum TSAb levels were measured at the Mayo Clinic using the Thyretain™ thyroid stimulating immunoglobulin (TSI) Reporter BioAssay (Quidel, San Diego, CA). In addition, a bioassay based on stimulation of cyclic adenosine monophosphate production in TSHR expressing Chinese hamster ovary cells was used at FIRS to measure serum TSAb activity and serum thyroid stimulating blocking activity (RSR™ TSAb and RSR™ TSBAb, respectively) (5).

Serum K1-70 concentrations were measured by enzyme-linked immunosorbent assay from RSR Ltd. Anti K1-70 antibody (ADA) levels were assessed using a Meso Scale Discovery platform. Thyroglobulin (TG) concentrations in serum were measured at the Mayo Clinic using a Beckman Access Thyroglobulin assay run on a UniCel DxI 800.

GO course

Soon after (within 22 days) the first dose of K1-70, the patient reported improvement in her eye symptoms. She was reassessed by a GO expert in August 2017 (after 4 months K1-70; Table 1), at which time her CAS decreased to 0–1/7 and exophthalmometry reduced from 21 to 19 mm bilaterally despite ongoing smoking. These improvements enabled eye surgery to correct her diplopia and restored her driving ability for the first time in several years (Fig. 1B). GO symptoms remained under control while on K1-70 therapy alone or in combination with lenvatinib. Her GO was stable, while lenvatinib was held for 4 months (July–October 2017; Table 1) due to intolerance, and continued to be stable when lenvatinib was restarted.

In April 2018, after 16 doses of K1-70 every 3 weeks, K1-70 was held before a whole-body ¹²³I scan and thyrogen-stimulated ¹³¹I therapy. During the break in K1-70 and before this third dose of ¹³¹I, her TSAb activity increased (Table 1) and the patient reported worsening eye symptoms, which required prednisone treatment (20 mg per day orally, decreasing to 5 mg/d). Three weeks after ¹³¹I, K1-70 treatment was resumed (June 2018) at 120 mg i.m., her GO improved rapidly and prednisone was discontinued. However, at this stage, serum concentrations of K1-70 were much lower compared with those seen with 120 mg i.m. dosing in the first few months of treatment. Consequently, in August 2018, the injection route was changed to i.v. (Table 1). The reasons why serum concentrations of K1-70 were not reaching the levels observed initially are not clear. Data from stability and compatibility studies available from the drug supplier did not indicate that K1-70 breakdown or adsorption to the drug delivery equipment could be responsible. There was no detectable serum ADA response after multiple (n = 14) K1-70 administrations. However, impaired absorption from the injection site could have been related to lenvatinib therapy. K1-70 serum concentrations were high and the corresponding TSAb activity lowest when the patient was receiving i.m. K1-70 while lenvatinib was on hold (July 2017–October 2017; Table 1).

When assessed by an ophthalmologist in October 2018, her eyes were white and quiet with CAS 0/7, exophthalmometry of 18mm bilaterally with no evidence of any clinical activity (Table 1). GO was stable, despite persistent smoking until January 2019 and thereafter until the patient's death in July 2019 due to acute cardiopulmonary collapse deemed unrelated to K1-70 therapy. In total, the patient received 2755 mg of K1-70 (1675 mg i.m. and 1080 mg i.v.) in 33 three weekly doses (19 i.m and 14 i.v.; Table 1). K1-70 was well-tolerated except for discomfort associated with i.m. injections. There were no adverse effects such as nausea, vomiting, headache, or allergic reactions following either i.m. or i.v. K1-70.

Serum K1-70 and TSAb

As doses of K1-70 increased, concentrations of serum K1-70 increased with the corresponding blocking of endogenous TSAb activity (from 2010% to 216% 150 days after starting the therapy; Fig. 1A). Also, serum thyroid stimulating blocking activity (TSBAb) activity increased accordingly (Fig. 1A). The decrease of TSI index and TSAb activity observed with increasing doses of K1-70 reflected net serum activities of the mixture of endogenous TSAb and administered K1-70. The excess of K1-70 in serum effectively blocked binding of the endogenous TSAb to the TSHR in the bioassays (1 –4,6).

FTC course

Despite multiple neck operations, thoracotomies, and high-dose ¹³¹I therapies for iodine-avid FTC (cumulative dose 500 mCi), her disease progressed rapidly. At the time of initiating K1-70 therapy, serum TG levels were 190 ng/mL on 20 mg per day of lenvatinib (Table 1). After 3 months on K1-70 (in July 2017), she reported feeling “very well” and requested a break in treatment with lenvatinib due to intolerance. Lenvatinib was held for 4 months while she continued on K1-70 therapy. Compared with prelenvatinib therapy, attenuation of tumor progression was observed on K1-70 alone and the Mayo Clinic quality-of-life score of 6/10 recorded on lenvatinib alone improved to 9/10. However, during this time, her tumors showed a mixed response. Some tumors were stable or showed slight or mild progression, while some were stable or showed slight regression on K1-70 alone. As some tumor foci showed progression and TG increased on K1-70 alone, lenvatinib was restarted in October 2017 although at reduced doses (10–14 mg/d) in combination with K1-70. Subsequently, the dose of lenvatinib was titrated up to a maximum of 24 mg per day in November 2018.

Her TG level of 800 ng/mL in December 2016, despite 250 mCi radioiodine 3 months prior, fell to 190 ng/mL in April 2017 (Table 1) after initiation of lenvatinib 20 mg daily. Following dose increments of K1-70 from April to September 2017, TG levels remained at about 200 ng/mL (Table 1) despite holding lenvatinib for 3 months due to toxicities, suggesting at least a transient effect of K1-70 itself on the FTC. This corresponded to observed stabilization of foci of disease on imaging described above (data not shown).

Due to the poor tolerability of lenvatinib and disease progression, salvage therapy with 300 mCi ¹³¹I was given as per Tumor Board discussion for persistent iodine-avid metastatic foci in May 2018. Unfortunately, although not unexpectedly, there was no evidence of tumor response. In November 2018, on lenvatinib 24 mg daily, 80 mg doses of K1-70 i.v., and detectable TSAb, her TG nadir was 1153 ng/mL (Table 1), suggesting combination therapy synergy on tumor control. Intolerance to lenvatinib continued with fatigue, nausea, vomiting, and loose stools. Due to dose-limiting toxicities leading to therapy interruptions and overall disease progression, treatment was switched in January 2019 to pazopanib, also dose adjusted for toxicities (600 mg/d). Her serum TG levels dropped to 694 ng/mL in June 2019 (Table 1) with stable structural disease. However, the patient developed pulmonary and respiratory symptoms, which responded initially to antibiotics but progressed to lung abscess/cavities leading to an acute cardiopulmonary collapse and death in July 2019.