The principle treatment of locally advanced DTC is surgery achieving R0 (no residual cancer) [10, 11]. In this case also, undoubtedly, the initial therapy should be surgery, including tracheal resection and reconstruction. After the operation, aggressive local recurrence and distant metastasis emerged, and as a standard treatment for such a case, a single dose of 100 mCi of radioactive iodine therapy was given with a negative diagnostic scan; therefore, instead of further radioactive iodine treatment, EBRT was primarily indicated for analgesic purpose, and combined systemic chemotherapy was followed. Because a lack of RAI uptake confers a poor prognosis, EBRT and systemic chemotherapy may be effective options. In our case, docetaxel was administered because its effectiveness was formerly suggested , and in fact, it was effective for a year; it controlled local recurrence in the neck, but lung metastasis was remarkably worsened. At that time, there was no further line of treatment because TKIs were unavailable; hence, terminal care may have been recommended.
Sorafenib then became commercially available. However, because of disease progression, it was discontinued. Lenvatinib then became available and was taken as the next line of treatment. The phase III SELECT study documented a significant improvement in the median PFS among patients treated with lenvatinib compared with those treated with the placebo (18.3 months vs. 3.6 months; HR, 0.21; 99% CI, 0.14–0.31; p < 0.001). Compared with other TKIs, lenvatinib is very effective because it has potency with regard to the inhibition of FGFR1–4, offering a potential opportunity to block a mechanism of resistance to the VEGF/VEGFR inhibitors.
Lenvatinib also has a direct oncogenic effect on the control of tumor cell proliferation by inhibiting RET and an effect on the tumor microenvironment by blocking FGFR [13, 14]. The SELECT study included prior anti-VEGF TKI treatments (sorafenib, 77%; sunitinib, 9%; pazopanib, 5%; and other, 9%), differing from the DECISION study, and these treatments were effective . In fact, we had no data regarding molecular-biological approaches, including western blotting analysis of pathologic specimens or new generation sequencing data of this patient. Nevertheless, we selected lenvatinib as an effective drug because, as Tahara et al.  reported, there should be no difference in drug efficacy because of genetic background, and lenvatinib was the last TKI available instead of sorafenib. In this case, lenvatinib was effective after the failure of sorafenib. To our knowledge, this is the first report of metastatic, RAI refractory, unresectable recurrent DTC in which possible multimodal treatments and other TKIs were ineffective, although a report existed indicating the effectiveness of lenvatinib as a fourth-line TKI for thyroid cancer .
For an effective use of TKIs, our institute regulations are reported to be very important and effective in introducing TKIs. Because of our adherence to these regulations, the use of lenvatinib has been maintained since a long time and has caused tumor shrinkage.
In contrast, lenvatinib caused tracheal fistula formation close to anastomosis. A history of EBRT is thought to increase the risk of fistulas . In our case, the fistula promptly closed after the cessation of lenvatinib, although there are some reports regarding the delayed healing of fistulas caused by lenvatinib, which is active against the FGFR family of tyrosine kinase receptors . The delayed healing of tissue may have been reversed by the cessation of lenvatinib .
Because the patient suffered from diabetic nephropathy, the control of diabetes apart from nutritional remedies—required the discontinuation of lenvatinib, which caused tumor recurrence. Therefore, it is suggested that in such patients, TKI does not kill malignant cells, but just stabilizes them, and that TKI should be given until tumor recurrence.
The timing of the use of TKI should be considered only when the benefits outweigh the risks, but as we cannot usually predict adverse events, patients’ requests might also be important . Although, it should be also emphasized that during the course of the treatment, lenvatinib was very effective, despite two instances of medicine cancellation and its subsequent resumption, which indicates that the effect can be restored by intermittently using the medicine, even if discontinued.
Possible future treatment strategies after the failure of lenvatinib are discussed below. First, lenvatinib and sorafenib are the only TKIs available in Japan; therefore, the already decreased dose of lenvatinib should be increased to certify effectiveness because Morelli and Puxeddu  have previously reported that increase in lenvatinib dose enabled disease control. Second, because efficacy of some BRAF or MEK inhibitors have been reported , they may also be available in the near future. Finally, immunotherapy is another option: Nivolumab plus Ipilimumab (NCT03246958) or WT1  vaccine may be promising.
To our knowledge, this is the first case report of recurrent PTC for which possible multimodal treatments were finally ineffective (PD) because of disease aggressiveness and the promising TKI lenvatinib was extremely effective (PR) for a long time (more than 1 year and 9 months). Moreover, it should also be noted that this case, which presented the perforation of the trachea invaded by the tumor, could be cured by a temporary cessation of the drug, and thereafter drug intake could be maintained by a proper management of adverse events or complications.