Redifferentiating Effect of Larotrectinib in NTRK -Rearranged Advanced Radioactive-Iodine Refractory Thyroid Cancer

Introduction

Ten percent of differentiated thyroid cancers will become metastatic despite adequate initial treatment (1). In this setting, a third of patients are cured by radioactive-iodine (RAI) therapy, more frequently for young patients with low burden tumor volume (2). For the other patients, cancer cells are RAI resistant, mainly due to the loss of iodine uptake. In this context, the standard therapy when systemic treatment is required relies on the Tyrosine Kinase Inhibitors lenvatinib (3,4), sorafenib (5), and cabozantinib (6) with significant adverse events (7).

Recently, the molecular landscape of thyroid tumors has been explored and led to the description of molecular alterations constitutively activating the mitogen-activated protein kinase (MAPK) pathway, with BRAF^V600E mutation present in ∼60% of papillary thyroid cancer, 33% of poorly differentiated thyroid cancer, and 45% of anaplastic thyroid cancer, whereas RAS mutations are described in 13% of the differentiated thyroid cancers (8,9). In this setting, BRAF and MEK inhibitors were shown to have both antitumor activity (10) and to induce tumor redifferentiation possibly restoring RAI uptake for patients with metastatic RAI-resistant thyroid cancer (11 –13).

Other and rarer molecular events exist in thyroid cancers. Two distinct recent observations of RAI reuptake under larotrectinib (a specific neurotrophic tyrosine receptor kinase [NTRK] inhibitor) suggested the possibility to extend this concept to NTRK rearrangements: one of these observations concerned an adult patient (14) and is further detailed in this study, and the other concerned a pediatric case (15). Furthermore, this concept may also apply for other rearrangements (15,16). Similar findings in other adult patients would confirm the potential therapeutic interest of a redifferentiation strategy for patients with NTRK-rearranged RAI-resistant metastatic thyroid cancer. We decided to explore whether redifferentiation could be observed in patients treated with larotrectinib for such diseases.

Patients

Since 2018, all patients treated in our institution for a RAI-resistant thyroid cancer have access as part of standard of care to a molecular profile of their tumor with next-generation sequencing to identify somatic mutations (Ion AmpliSeq^® covering various hotspots including genes of interest for thyroid cancers such as BRAF, NRAS, KRAS, and HRAS). If none is identified, tumors are screened for molecular rearrangements (RNA next-generation sequencing using an Archer Fusionplex^® Lung Panel, including ALK, RET, BRAF, and NTRK1-2-3 fusions).

During this period, three patients for who an NTRK rearrangement was identified were eligible to receive larotrectinib in a national access program for a metastatic disease and underwent diagnostic scintigraphy to assess redifferentiation. All patients provided written informed consent for the use of their data, and the study was conducted in accordance with the amended Declaration of Helsinki and relevant local rules (local committee approval CLEP No. AAA-2022-08002).

Patients' characteristics and treatment sequences are presented in Table 1.

Patient A has already been briefly described elsewhere (14); she was diagnosed with papillary thyroid cancer 35 years before larotrectinib treatment and had received multiple RAI treatments, the last one 14 years ago for the emergence of a pulmonary micrometastatic disease. More recently, the cancer had become active with symptomatic and radiographic disease progression and lenvatinib was started leading to a partial response according to Response Evaluation Criteria In Solid Tumors (RECIST).

No RAI uptake was observed under lenvatinib. Poor tolerance of lenvatinib despite dose adaptation led to treatment interruption. Since an EML4-NTRK3 rearrangement had been identified, larotrectinib was initiated with good tolerance except for Common Terminology Criteria for Adverse Events (CTCAE) grade 1 withdrawal pain (defined as pain experienced during temporary or permanent discontinuation of a treatment) and CTCAE grade 2 orthostatic dizziness effectively treated with fludrocortisone, both events being known on-target adverse events of NTRK inhibition (17).

Patient B had a history of papillary thyroid cancer with immediate RAI resistance of the pulmonary micrometastatic disease. She received lenvatinib for two years with partial response according to RECIST but the cumulative toxicity despite dose adjustment led to a treatment interruption with rapid disease progression. Since an ETV6-NTRK3 rearrangement had been identified, larotrectinib was started with good tolerance except for CTCAE grade 1 weight gain, which is also a known on-target adverse event of TRK inhibition (17).

Patient C was newly diagnosed with oxyphilic cell papillary thyroid carcinoma with widespread metastatic disease at onset. Total thyroidectomy was performed, and postoperative diagnostic scintigraphy showed only cervical RAI uptake, with limited node uptake, but no uptake in the metastatic disease. The patient then suffered from a pathological femoral bone fracture that justified orthopedic surgery. In the context of a rapidly evolving symptomatic disease with contraindication to a Vascular Endothelial Growth Factor targeted treatment and since a TPM3-NTRK1 rearrangement had been identified, larotrectinib was started with good tolerance.

The redifferentiation evaluation used for each patient was similar except for the duration of larotrectinib before RAI diagnostic scintigraphy that varied from 20 to 30 days. Larotrectinib was given orally at the dose of 100 mg twice daily every day and continued until disease progression or unacceptable toxicity. Whole body scans were performed on an NM870 SPECT-CT gamma camera (GE) equipped with a dedicated highly sensitive (73 cps/MBq for ¹³¹I) 5/8″ thick crystal. Patients were prepared with a regular recombinant thyrotropin stimulation (0.9 mg × 2). The used ¹³¹I activity level is higher than that routinely used for imaging purpose and provides highly sensitive quality images with our system, even in nonpreviously ablated patients.

Figure 1 illustrates the different scans of the three patients before and under larotrectinib. For Patient A [as previously described (14)] and Patient B larotrectinib-induced redifferentiation was observed, whereas for Patient C no uptake was observed in the metastatic lesions previously non-RAI-avid (Supplementary Fig. S1). Noteworthy, the three-month tumor evaluation showed partial radiological response according to RECIST for all three patients and including metabolic response on all lesions for Patient C on the positron emission tomography imaging.

OPEN IN VIEWER

FIG. 1.

Post-RAI whole body scans and PET obtained before and after larotrectinib treatment. The first column shows the most recent 18FDG PET imaging before larotrectinib initiation, the second column shows the most recent post-RAI scintigraphy before larotrectinib initiation, and the third column shows the post-RAI scintigraphy under larotrectinib. For Patient A, diffuse reuptake of RAI is seen in pulmonary lesions but neither in the metastatic lymph nodes nor in the fourth lumbar vertebra lesion, 18FDG avid on PET. For Patient B, RAI reuptake can be seen in all known pulmonary and left cervical lymph node metastasis. For Patient C, only cervical lymph node shows RAI uptake since baseline, but no RAI reuptake is induced by larotrectinib while multiple 18FDG avid metastatic lesions are present on PET. 18FDG, 18 fluorodeoxyglucose; PET, positron emission tomography; RAI, radioactive-iodine.

Discussion

To the best of our knowledge, we report the first series of patients with RAI-resistant metastatic thyroid cancer with NTRK rearrangement for whom larotrectinib-enhanced RAI uptake was investigated. Our main finding is that redifferentiation was observed for two patients out of three with no new or unexpected safety signal. The potential mechanism of action for this redifferentiating effect of larotrectinib could be quite similar to that of MAPK inhibitors (15). In this article by Lee et al., in vitro experiments showed that larotrectinib-induced restoration of RAI uptake was mediated by sodium/iodide symporter re-expression. The redifferentiating effect also appears not to be limited to NTRK rearrangements but may also concern RET rearrangements (15,16).

For RAI-resistant thyroid cancers with point mutations, the redifferentiating effect induced by molecularly targeted agents also appears inconstant and not linked to the ability of the drug to induce tumor response. For thyroid cancers harboring BRAF mutations treated with BRAF inhibitors, possibly associated to MEK inhibitors, redifferentiation can be observed in ∼60% of the cases (11,12,18,19). For these tumors, the absence of redifferentiation upon MAPK blockade may result from mutations in SWI/SNF genes resulting in loss of chromatin accessibility (20).

Our series also shows different clinical courses of NTRK-rearranged thyroid cancers. We highlight that larotrectinib-induced redifferentiation may occur in tumors with acquired RAI resistance (Patient A) but also in tumors with primary RAI refractoriness (Patient B). Noteworthy, Patient A and B had a slowly evolving disease while Patient C presented with an aggressive rapidly evolving disease, showing heterogeneity in the clinical presentation of NTRK-rearranged diseases. It can also be noted that despite a clear clinical and radiological benefit of larotrectinib for Patient C, no uptake was identified demonstrating a difference between the antitumoral and the redifferentiating effect.

The therapeutic implication of our findings is still unknown since we only report on the ability of larotrectinib to enhance RAI uptake. Future studies with patients receiving RAI at therapeutic dose will allow answering the question of the efficacy of the redifferentiation strategy and will help distinguishing the therapeutic effect of NTRK inhibition from that of RAI.

In conclusion, we demonstrate that a redifferentiation process can be observed with larotrectinib for NTRK-rearranged RAI-resistant thyroid cancers.