The combination of surgery, radioactive iodine (RAI), and L-thyroxine therapy is effective treatment for patients with differentiated thyroid cancer (DTC), which accounts for some 95% of all thyroid cancers.
1-4Distant metastases, however, can occur in approximately 23% of patients, and are the main cause of death from DTC.3,5Moreover, a subset of these metastatic patients will develop RAI-refractory disease, which has a poor prognosis.6In one study of 444 patients with DTC, those with metastatic RAI-refractory DTC (n = 132) had an approximate survival of between 2.5 years and 3.5 years, and nearly 80% of these patients died of thyroid cancer.6Standard cytotoxic chemotherapy (ie, doxorubicin) is of limited benefit for these patients.3
Studies have implicated vascular endothelial growth factor (VEGF) and the BRAF, RAS, and RET molecular pathways in DTC.7,8Sorafenib is an orally administered multikinase inhibitor with inhibitory activity at all 3 known VEGF receptors (VEGFRs 1-3), as well as BRAF, RET, and platelet-derived growth factor receptor (PDGFR) beta, and earlier studies have shown activity of this agent in advanced RAI-refractory DTC.4,9,10The DECISION trial was designed to evaluate the efficacy and safety of sorafenib in patients with locally advanced or metastatic RAI-refractory DTC in a randomized, placebocontrolled, phase III study.5Marcia S. Brose MD, PhD, assistant professor of otorhinolaryngology head and neck surgery at the University of Pennsylvania and lead author of the study, noted that before the DECISION study, no treatment options existed for these patients besides palliative care.
Progression within the past 14 months per RECIST; at least 1 measurable lesion by CT or MRI as per RECIST
Presence of at least 1 target lesion without RAI uptake; or patients progressing after 1 RAI treatment within the past 16 months; or; progressed after 2 RAI treatments within 16 months of each other (with last treatment >16 months ago) or received a cumulative RAI activity of at least 22.3 GBq (>600 mCi)
CT indicates computed tomography; MRI, magnetic resonance imaging; OR, overall response; RAI, radioactive iodine; RECIST, Response Evaluation Criteria in Solid Tumors
“The only drug approved for this indication was doxorubicin, and that was approved in 1974, before we had CT scans.” Furthermore, she emphasized that once CT scans became available, “We realized it (doxorubicin) really did not workthere was only about a 5% response rate, and it was toxic.” Consequently, she said, doxorubicin had generally not been used in this setting for 10 to 15 years, and accordingly, the standard of care for individuals with progressive, RAI-refractory DTC was essentially palliation. As a patient’s disease progressed, “We tried to palliate as much as possible using radiation or surgery,” Brose said.The key entry criteria for the DECISION trial are summarized inTable 1. Briefly, adult patients were included if they were 18 years of age or older, with locally advanced or RAI-refractory DTC, Eastern Cooperative Oncology Group (ECOG) performance status 0-2, with adequate bone marrow, liver, and renal function.5Commenting on the entry criteria for the trial, Brose emphasized the importance of first, identifying those patients who had failed RAI therapy, and second, identifying those who had locally recurrent and/or metastatic diseasein other words, a patient’s disease was required to be progressing. “Some patients have a honeymoon period, where they stop taking up radioiodine, but their disease is not growing—these people can be observed for a time and are not in need of therapy,” she said. Brose also stressed that the cutoff of 1 cm (per standard RECIST [Response Evaluation Criteria in Solid Tumors] criteria) is a clinically relevant cutoff, because most patients with disease less than 1 cm can likely be observed for a period before starting therapy.
“I’m a strong proponent of identifying patient populations in need for clinical trials. In the DECISION trial, in the placebo arm, the progression-free survival (PFS) was 5.8 months, so these were patients that were clearly in need of therapy. The criteria for the trial, that is, RAI-refractory disease that is growing, and (is) no less than 1 cm, did a good job in selecting the patient population in DECISION,” Brose stated.Patients in DECISION were randomly assigned (1:1) to either sorafenib (2 × 200-mg tablets, administered twice daily, 12 hours apart, without food; total daily dose, 800 mg) or matching placebo, until progression, unacceptable toxicity, noncompliance, or consent withdrawal.5Dose interruption, sequential dose reduction, and re-escalation of dose were permitted according to specified criteria to manage adverse events (AEs). The primary end point was PFS, as assessed through modified RECIST criteria, every 8 weeks. Overall survival (OS), time to progression (TTP), and objective response rate (ORR) were included among the secondary endpoints. In an effort to identify potentially predictive, prognostic, or pharmacodynamic factors, exploratory biomarker analyses were also conducted using formalin-fixed, paraffin-embedded biopsy material (primary or metastatic sites) from consenting patients.5
HR indicates hazard ratio; PFS, progression-free survival; ORR, overall response rate; TTP, time to progression.
A total of 419 patients were randomly allocated to sorafenib (n = 209) or to placebo (n = 210), and baseline demographics were well balanced between the treatment arms. More than 95% of patients had distant metastases, with the 3 most common sites being lung, lymph nodes, and bone, respectively. Results for the primary endpoint showed a significant improvement in PFS for patients in the sorafenib arm (Figure 1), with a hazard ratio (HR) of 0.59 (95% confidence interval: 0.45-0.76;P<.0001), corresponding to a 41% relative reduction in the risk of progression or death. Significant improvement in TTP and ORR was also observed with sorafenib relative to placebo (Figure 1).5Improvement in PFS with sorafenib was consistent among prespecified subgroups in exploratory analyses. There was no significant benefit in OS (HR = 0.80;P= 0.14), however, this endpoint was likely confounded by the trial-permitted crossover from placebo to open-label sorafenib upon progression.
Commenting on these findings, Brose noted that the study met the primary endpoint, extending PFS by 5 months (Figure 1). “For a patient going in and out of the hospital every time they have a progression, I think that’s clinically significant. This is progress we have not had in over 40 years.”Nearly all patients (98.6%) receiving sorafenib reported AEs (Table 2) during the double-blinded phase of DECISION (vs 87.6% of patients on placebo). Most of these events were seen early in the course of treatment and were of grade 1 or 2; hand-foot skin reaction (HFSR), diarrhea, alopecia, and rash/desquamation were most common (Table 2). Of note, an increase in thyroid-stimulating hormone (TSH) of >0.5 mIU/L occurred in 33.3% of patients in the sorafenib arm, and 18.8% of patients reported hypocalcemia.5
“The most important message about AEs is that they are vastly manageable,” Brose said, noting that, in most cases these could be addressed with over-the-counter medications or with titration of the dose. “It is important to remember that the trial started patients on the full dose (of sorafenib) and then worked them down if they could not tolerate it; a lot of patients got full dose or three-quarter dose, so it cannot be guaranteed, for example, if one starts every patient at half-dose, that you will get the same results,” Brose cautioned.
Sorafenib (n = 207)
Placebo (n = 209)
Gr 3/Gr 4 (%)
Gr 3/Gr 4 (%)
20.3 / -
0 / -
5.3 / 0.5
1.0 / 0
- / -
- / -
4.8 / 0
0 / 0
5.3 / 0.5
1.4 / 0
5.8 / -
1.0 / -
9.7 / 0
2.4 / 0
HFSR indicates hand-foot skin reaction.
Brose noted that there were some AEs specific to thyroid cancer with this drug; when considering diarrhea, for example, she emphasized that often these patients are borderline hypoparathyroid, and that calcium should be monitored and corrected if needed. Lastly, she noted that patients with DTC are usually kept at a suppressive dose of thyroid hormone, in order to reduce their TSH levels, and with increasing time on sorafenib, about onethird of the patients required increased doses of thyroid hormone to maintain suppression of their TSH.
Strategies for the management of specific AEs with sorafenib, including HFSR, upper and lower GI distress (eg, diarrhea), hypertension, and cardiovascular issues, are extensively detailed, from the clinician’s perspective, in a recent publication.11Recognition and prompt, proactive management of AEs allow for a maximal therapeutic benefit by enabling patients to stay on their prescribed dose as long as possible.11While limited by small sample size (only 61.4% of the study population), exploratory analyses evaluating PFS according toBRAFandRASmutational status demonstrated no independent prognostic or predictive value for these markers in relation to sorafenib benefit. Notably, median thyroglobulin levels increased in patients in the placebo arm, while they initially decreased in patients on sorafenib, suggesting that such changes in this marker could be a result of disease progression. Consistent with this notion was the finding that, while levels were stable in those with radiological stable disease, patients with partial responses had the greatest decreases in thyroglobulin, and, conversely those with an initial drop and subsequent increase showed radiological progression. Overall, however, the extent to which serum thyroglobulin could be used to evaluate treatment with sorafenib remains to be further investigated. “Thyroglobulin levels are still kind of a work in progress; we are definitely studying it, but at this point, no treatment decisions should be made based on the thyroglobulin,” Brose cautioned, noting that the drug itself probably changes, to some extent, how the thyroglobulin is produced.
“The initial drops and increases may have nothing to do with the tumor burden. That said, there may be some signals with thyroglobulin that can help lead people to maybe have closer surveillance,” she said.The pivotal phase III findings from DECISION identified a long-awaited treatment option for patients with RAI-refractory DTC. Adverse events with sorafenib were generally manageable, and consistent with the drug’s previously established safety profile. While not predictive of treatment benefit, preliminary findings with thyroglobulin levels point to its potential use as a pharmacodynamic marker.