Real-world clinical profile, treatment patterns and patient-reported outcomes for thyroid cancer in Japan

The incidence of thyroid cancer has steadily increased in recent decades. From 1990 to 2017, incident cases of thyroid cancer rose by 169%, and deaths by 87%, with an age-standardized estimated annual percentage change of 1.59 [1]. In Japan, in 2020, thyroid cancer was ranked 13th by incidence and 22nd by mortality among all cancers [2]. Thyroid cancers are classified into three main histological types: differentiated thyroid cancer (DTC); medullary thyroid cancer (MTC) and anaplastic thyroid cancer (ATC). DTC accounts for more than 95% of cases, of which papillary thyroid cancer (PTC) is the most common subtype [3]. In contrast, other forms of thyroid cancer, such as MTC and ATC, account for only a small fraction of cases (1–2% and <1%, respectively) [3].

Rearranged during transfection (RET) gene fusions are present in 10–20% of PTC [4], and mutations in the proto-oncogene RET are responsible for most cases of MTC [3]. MTC may be either somatic or hereditary in origin, and RET gene mutation analysis is recommended for all patients with MTC [5]. RET mutations are detected by mutational analysis in almost all hereditary cases (>98%), usually in exons 10, 11, or 13–16 [5].

Treatment for thyroid cancer mainly consists of surgery, radioactive iodine (for DTC only), and molecular targeted therapy [3,5,6]. Systemic therapies are an important treatment option for patients with advanced thyroid cancer, and among these multikinase inhibitors (MKI) are frequently used. In Japan, lenvatinib is approved in DTC, MTC and ATC; sorafenib is approved in DTC and MTC, while vandetanib is approved for MTC only [5]. Since these drugs are associated with significant adverse events including hypertension, diarrhea and weight loss, and are administered in the long-term, careful assessment of the risk-benefit profile must be made based on tumor load, symptoms, location of metastases and clinical status prior to their initiation [6]. In the National Comprehensive Cancer Network guidelines, MKIs are considered as preferred treatment options for progressive and/or symptomatic disease [7]. In the Japan Association of Endocrine Surgeons (JAES) guidelines [5], for asymptomatic patients active surveillance is recommended, and treatment should not be started unconditionally for all patients who have lesions with rapid growth or symptoms, but the timing and initiation of therapy should be determined by considering the benefits and risks associated with the therapy as well as the patients' general performance status. However, the optimum duration of surveillance prior to initiation of MKI is unknown in clinical practice, and the timing for initiation of an MKI is also debated [8]. In addition, although these MKIs contribute to tumor response and progression-free survival prolongation [9], there is limited published research on the impact of health-related quality of life (HRQoL) in patients with thyroid cancer during MKI treatment.

Limited information has been published, especially at multiple centers that provide an overview of the real-world thyroid cancer management landscape in Japan, including the use of MKIs, genetic tests and the hurdles or opportunities toward better thyroid cancer management, also considering that it is managed by different specialists and types of hospitals. Herein, with the aim of better understanding current practice, we present the clinical profile, treatment patterns and patient-reported outcomes (PROs) for advanced PTC and MTC in Japan.

Materials & methods

Study design & data source

Data were derived from the Adelphi Thyroid Cancer Disease Specific Programme (DSP™) that was carried out in 2020 in several countries, including Japan. These are large, multinational surveys that aim to understand current trends in disease management, disease-burden impact, and the impact of treatment in a real-world setting. Physicians from Japan were asked to complete a patient-record form (PRF) for up to the next five consecutive patients who they saw during routine care and who met the study inclusion criteria. Each patient for whom the physician completed a form was also asked to complete a patient self-completion form (PSC) on a voluntary basis and provide informed consent to participate. A physician survey was also given to participating physicians that contained information on routine practice.

A complete description of the methods of the survey has been previously published and validated [10–12], and the main points are outlined in this section. Briefly, the survey method collects both qualitative and quantitative data from eligible physicians and their patients. Since it is a cross-sectional survey with retrospective data collection which had no pre-defined, interventional laboratory testing or treatment regimens, results provide real-world data, reflective of clinical practice. Although data were collected across several countries, the current study conducted in Japan can be considered an independent country-specific study.

Ethical considerations

Using a check box, patients provided informed consent for use of their anonymized and aggregated data for research and publication in scientific journals. Data were collected in such a way that patients and physicians could not be identified directly; all data were aggregated and de-identified before receipt.

This research also obtained ethics approval from the Western Institutional Review Board, study protocol number #AG8757.

Data collection was undertaken in line with European Pharmaceutical Marketing Research Association guidelines [13] and as such it does not require ethics committee approval. Each survey was performed in full accordance with relevant legislation at the time of data collection, including the Health Information Technology for Economic and Clinical Health Act legislation [14].

Participating physicians & patients

Physicians were eligible to participate in this survey if they see ≥2 patients with MTC or PTC in a typical 12-month period and are responsible for treatment decisions and management of these patients. Patients were eligible for inclusion if they were over the age of 18 years at data collection, had a physician-confirmed diagnosis of locally advanced or metastatic MTC or PTC at data collection, and visited the physician. No tests, treatments, or investigations were performed as part of the survey. Participating physicians were reimbursed upon survey completion according to fair market research rates. Patients were not compensated for participation.

Study variables

Physicians completed PRFs that included general information on patient demographics and referral, diagnosis, life expectancy at diagnosis, disease status and Eastern Cooperative Oncology Group (ECOG) performance status at data capture, details on prior treatments received, biomarker/genetic testing and concomitant conditions. Physicians also completed a survey that captured information on their routine practice including specialty, patient workload and management, work setting and questions related to testing for RET alterations including perceived barriers to RET testing. PSCs were completed by the patient independently and returned in a sealed envelope to ensure that the responses were confidential. The PSC captured information on demographics, caregiver utilization, care management, goals for treatment, treatment satisfaction and PROs. PROs were obtained using three validated instruments: the 5-level EuroQol 5 Dimension questionnaire ([EQ-5D-5L] subdomains: mobility, self-care, usual activities, pain/discomfort and anxiety/depression); the Work Productivity and Activity Impairment (WPAI) questionnaire; and the Functional Assessment of Cancer Therapy-General (FACT-G) questionnaire. Health utility index scores were generated using specific tariffs for Japan, with lower scores reflecting poorer HRQoL [15]. Health utility index scores range from <0 (worse than death) to 1 (perfect health) [16]. The WPAI is a six-item questionnaire that measures the impact of disease on work productivity and daily activities over the past week, and the score is expressed as a percentage, with higher scores indicating a greater impact [17]. FACT-G is a 27-item questionnaire designed to measure four domains of HRQoL in cancer patients: physical; social; emotional and functional well-being. The FACT-G total score ranges from 0 to 108 points, with higher scores indicating better HRQoL [18].

Statistical analyses

Results are presented separately for patients with MTC and those with PTC. Descriptive statistics are provided for demographics and disease characteristics. Continuous variables are mainly described using mean and standard deviation (SD), and/or median or range where it is common to do so; there was no differentiation between data with normal and non-normal distribution in our analysis. Categorical variables are reported as the frequency and percentage within each category. As this was a descriptive study with pairwise case deletion used in data analysis, missing data were not imputed; this was deemed unlikely to alter conclusions. The number of responses for the analyses was variable and is thus reported separately for each analysis.

Results

Patient & physician characteristics

Characteristics of patients with MTC (n = 56) and PTC (n = 106) are detailed in Table 1. The median age was similar in both cohorts (67.5 and 68.5 years, respectively). The proportion of women was higher in PTC compared with MTC (80 vs 46%). In all, 61% of patients with MTC and 58% of those with PTC had metastatic disease at data capture. The proportion of patients who had received previous treatment or were currently receiving treatment was ≥50% for MTC and ≥70% for PTC. Treatment comprised one or more of surgery, radiotherapy, drug therapy and best supportive care. ECOG performance status scores were numerically worse overall at data capture than at diagnosis (median time from initial diagnosis to data capture of 27.4 months for MTC and 39.6 months for PTC), suggesting a deterioration in overall health status. Life expectancy was ≥5 years in the majority of patients with both MTC (57%) and PTC (53%).

The characteristics of the 36 physicians participating in the survey are listed in Supplementary Table 1. The types of centers in which the physicians practiced varied between cancer centers and university, general and community hospitals, with the highest proportion practicing in the latter (44%). A variety of medical specialties were also declared; most were medical oncologists (28%), internal medicine specialists (25%) and endocrinologists/diabetologists (19%). There was also a large range in the number of MTC/PTC patients seen in the past 12 months, though the median number of patients seen was low, especially for MTC (3.5 for MTC; 7.5 for PTC). Nearly half (47%) of physicians said that COVID-19 had not substantially affected patient management.

Testing for RET alterations

Compared with PTC, a larger percentage of patients with MTC had RET testing (24% [25/106] vs 55% [31/56], respectively; Table 2). Among those tested, 16% (4/25) of PTC were positive for RET fusions; 42% (13/31) of cases of MTC were positive for a RET germline mutation. Of the 13 patients with MTC who were positive for a RET germline mutation, ten (77%) were offered genetic counseling, and among these six had received it by the time of data capture. The physicians cited a large variety of challenges for testing RET alterations (Supplementary Table 2). Specific therapies not available/no therapeutic relevance/no influence for patient management was a major reason not to conduct a RET test in MTC (38%) and in PTC (53%). Testing for RET mutation and RET fusion for MTC and PTC, respectively, was much less frequent among physicians practicing at general hospitals compared with those at university and community hospitals (Supplementary Table 3).

Systemic antitumor therapies

Among the patients enrolled in this study, 48% (27/56) of MTC and 48% (51/106) of PTC patients had received, or were receiving at time of data collection, systemic antitumor therapies (Table 3). Patients on active drug treatment reported symptoms more frequently (7 and 23% asymptomatic in MTC and PTC, respectively) than those without active treatment (41 and 44% asymptomatic in MTC and PTC, respectively). Considering the antitumor therapies received at first-line, most patients with MTC and PTC were administered MKIs, including lenvatinib (26%) and vandetanib (22%) in MTC and lenvatinib (51%) and sorafenib (27%) in PTC (Table 3). On the other hand, some patients received cytotoxic chemotherapy at first-line including 15% of MTC treated with 5-FU and 10% of PTC with cisplatin. The mean (SD) time from initial diagnosis to initiation of a MKI was 23.6 (26.9) months for MTC and 29.3 (45.1) months for PTC. Among the patients who experienced recurrence/metastasis, the mean (SD) time from recurrence/metastasis to initiation of a MKI was 34.5 (28.5) months for MTC and 18.5 (32.6) months for PTC.

Table 4 lists the side effects reported with MKI among the patients who received MKI as current or most recent treatment. For patients with MTC, the most frequent were loss of appetite and decreased appetite, followed by constipation, diarrhea, hypertension, nausea, fatigue and QT prolongation (all with the same frequency). For patients with PTC, the most frequent were loss of appetite, hand foot syndrome and nausea, followed by decreased appetite, hypertension and fatigue (all with the same frequency).

Perspectives on systemic antitumor therapies

The physicians cited a wide range of reasons for prescribing a MKI (Supplementary Table 4). For both MTC and PTC, the most frequently cited were duration of response (12 and 39%, respectively), benefit for overall survival (12 and 37%, respectively) and progression-free survival (29 and 24%, respectively) and recommended by national guidelines (24 and 55%, respectively). A convenient dosing regimen and method of administration were also mentioned for PTC by up to approximately 24% of physicians. For MTC, suitable for patients with RET gene mutation was also cited by 35% of physicians.

Among the areas considered necessary for improvement for the current/most recently received MKI, a low incidence of serious events/severe side effects, overall survival benefit and improves patient's quality of life were most frequently cited for patients with MTC (18% for each category; Supplementary Table 5). For patients with PTC, the most frequently cited areas were low incidence of serious events/severe side effects (17%), positive cost benefit ratio (14%) and duration of response (11%).

Concerning the goals of treatment other than cure, the two most frequently mentioned for MTC and PTC patients were improved overall survival, progression-free survival and disease-free survival benefit (38 and 28%, respectively) and improved quality of life (18 and 31%, respectively; Supplementary Table 6).

Impact on HRQoL & daily life

The mean (SD) EQ-5D-5L index scores at data capture were 0.76 (0.20) and 0.79 (0.14) for patients with MTC and PTC, respectively (Table 5). Mean (SD) FACT-G scores at data capture were 63.2 (18.9) and 65.4 (16.2) for patients with MTC and PTC, respectively (Table 5). WPAI mean (SD) activity impairment was 45.0% (29.5) for patients with MTC and 37.7% (25.5) for PTC (Table 5). Results for each domain of EQ-5D-5L are shown in Supplementary Table 7.

Discussion

The present analysis provides a comprehensive, real-world snapshot of the clinical profile, treatment patterns and PROs for advanced thyroid cancer in routine clinical practice prior to the introduction of RET inhibitor therapy in Japan. Testing for RET alterations varied depending on the type of hospital/center. Common reasons cited by physicians for not testing included a lack of therapeutic relevance for patient management and unavailability of specific targeted therapy at the time of data collection. MKIs were the most frequently prescribed systemic therapy, and there was often a substantial lag-time between initial diagnosis or recurrence/metastasis and initiation of such treatment. The timing of MKI initiation was also highly variable. Adverse events with MKI therapy were frequently cited as an area for improvement in the physician survey. Additionally, PROs using validated instruments (e.g. EQ-5D-5L) showed patients with MTC and PTC had substantial disease/treatment burden.

Overall, the cohorts of patients with MTC and PTC appear to be largely similar to a previously described real-world population of patients with advanced thyroid cancer undergoing treatment with MKIs in Japan [19]. In that post-marketing observational study, the median age of patients with MTC was 63 years, with a large age range of 43 to 85 years [19]. This is similar to the present cohort of patients with MTC, which also had a wide age range. The sites of distant metastasis in the present study were also similar to those reported by the previous study, albeit at lower percentages overall in the present cohorts. This may be because of the limitation of assessing medical records retrospectively, wherein only the information needed for daily clinical practice is recorded.

The request for RET germline testing was not entirely in line with current guidelines. In MTC, the American Thyroid Association and JAES guidelines both recommend genetic testing for all cases of MTC to detect a germline RET mutation, although adherence to these guidelines is not 100% and needs to be improved [5,20]. Nearly half of the MTC patients in our study did not undergo testing, which is broadly similar to real-world data from the USA showing that more than a third of patients with advanced MTC were not tested for RET mutation as recommended by national guidelines [21]. Some of the challenges cited for RET testing in our study included high costs and lack of in-house testing facility, despite availability in Japan of outsourced RET testing. It is possible that adherence to guidelines will improve over time, as has been shown in studies examining trends in adherence to surgical guidance in thyroid cancer [22,23]. Also noteworthy is that our study was conducted prior to the introduction of the RET inhibitor selpercatinib in Japan; availability of biomarker-targeted therapy options can facilitate testing for the respective biomarker.

Among MTC patients with a positive RET mutation test result, 23% were not offered genetic counseling. This is a highly relevant issue for both patients and their families when considering surveillance and treatment. One of the reasons for this is related to the lack of sufficient genetic counselors in individual institutions in Japan, despite the fact that insurance billing through counseling is available [24]. This can be considered as an area where improvement is needed, especially in the era of precision medicine [24]. Indeed herein, the lack of available patient counseling was held to be a major challenge for RET germline mutation testing.

A personalized approach, enabled by biomarker testing, is warranted considering that the genetic profile of the tumor is crucial in predicting response to therapy [25,26]. For some of the patients with MTC in this study, MKIs were prescribed as first-line therapy because physicians thought that MKI was suitable for patients with RET gene mutation, indicating that testing and targeting RET was considered by a proportion of prescribers. In PTC, the role of RET fusion testing has traditionally been controversial [27,28]. In line with this, specific therapies not available or with no therapeutic relevance/impact on management was a major reason why physicians did not test PTC patients for RET fusion in this study. However, following the collection of data for this study, tumor tissue-based RET fusion and RET mutation tests have been approved and are now reimbursed in Japan. The results of these tests can be used to prescribe RET inhibitors, which are generally associated with favorable efficacy and a good safety profile [29,30]. Clinical usage and the number of tumor tissue-based RET fusion and RET mutation tests is thus expected to increase in the future. Regarding the anticancer therapies prescribed at first-line, MKIs were the main systemic therapy, which corresponds to recommendations in current national guidelines [5,20]. However, it was unexpected that some patients with MTC or PTC received cytotoxic chemotherapy as first-line treatment, although the percentage was low. In the JAES guidelines, it is recommended to not use cytotoxic anticancer chemotherapy for advanced/recurrent DTC [5]. MKIs are recommended in first line for radioiodine-refractory advanced/recurrent DTC and advanced/recurrent MTC [5], and these were the most frequently prescribed systemic therapy. Regarding the choice of MKI, lenvatinib (51%) and sorafenib (27%) were used in patients with PTC and lenvatinib (26%) and vandetanib (22%) were prescribed for MTC. The use of lenvatinib and sorafenib in PTC is consistent with the JAES guideline recommendations, which note that response rates are higher with lenvatinib than sorafenib [5]. The higher percentage of lenvatinib use compared with vandetanib in MTC was unexpected, since vandetanib is given a stronger recommendation than lenvatinib in prescribing guidelines [5]. However, the prescribing pattern might be influenced by other factors, such as experience and familiarity with lenvatinib in PTC. In fact, herein, familiarity with MKI treatment was one of the most cited reasons to prescribe a MKI.

For the optimal timing of initiation of a MKI, most guidelines recommend starting a MKI only in patients with symptomatic, rapidly progressing disease, or life-threatening lesions, even if there is lack of consensus over the precise criteria to use to prescribe a MKI [31]. In addition, it is not clear in routine practice when a MKI should be initiated for patients with recurrent/metastatic disease. Herein, it took almost 3 years in patients with MTC and 1.5 years in those with PTC from disease recurrence or diagnosis of metastatic disease before MKI treatment was initiated. Although patients with PTC would have undergone previous radioactive iodine therapy, it is possible that patients might have been followed by active surveillance in the absence of symptoms and disease progression before MKI treatment. It is difficult to judge whether this duration is most appropriate for patients. However, it was also evident that performance status had deteriorated at data capture compared with initial diagnosis. This is reflected by the data showing ‘improved quality of life’ as a relatively common current treatment goal other than cure. Lower performance scores are associated with improved clinical outcomes following MKI treatment, which suggests that starting therapy before the patient's performance deteriorates may be beneficial [32]. Earlier initiation of a MKI may thus have the potential to improve outcomes. Additionally, in a recently published study in patients with radioiodine-refractory DTC receiving lenvatinib, those with a lower tumor burden had prolonged overall survival compared with those with a higher tumor burden, further implying that early initiation of lenvatinib may improve outcomes in these patients [33]. The precise timing of MKI initiation needs to be optimized by future clinical studies.

A lower incidence of severe side effects was also reported in this study as an attribute of MKI therapies requiring improvement. Proactive management of adverse events is thus of value in order to optimize therapy and achieve this goal [34]. Loss of appetite/decreased appetite, followed by nausea, hypertension and fatigue, in both MTC and PTC, in addition to hand foot syndrome in PTC, were the most frequent side effects noted, which is largely consistent with those reported in clinical trials with MKIs [35–37]. Hypertension and decreased weight have been reported in around 70% of patients being treated with lenvatinib in real-life settings [38], similar to what is reported in the pivotal phase 3 trial [36]. In a post-marketing observational study in Japan, hypertension of any grade was reported in 79.4% of patients with DTC and in 64.3% of those with MTC; hypertension was grade ≥3 in 60.2% of those with DTC and 39.5% of patients with MTC [19]. In our study, patients receiving active drug treatment more frequently reported that they experienced symptoms compared with those without active treatment, suggesting that some symptoms were related to drug therapy, although it is also possible that the prior presence of symptoms was the reason that systemic therapy was initiated. In any case, improvement of symptoms through better disease treatment and management, including but not limited to drug therapy, should be a prioritized need, considering the treatment goals mentioned before. Biomarker testing to identify patients that could be treated with specific inhibitors with a favorable safety profile should be encouraged as a measure to improve disease management.

The results of the EQ-5D-5L index score and WPAI indicated that patients had impaired HRQoL and functioning in work/daily activities. Of note, the EQ-5D-5L index score was lower than the mean score of 0.91 for the general population with similar age in Japan [39], indicating that MTC and PTC patients had at least moderate disease and treatment burden. This finding is in agreement with a study in the UK in which EQ-5D utility health scores in patients with thyroid cancer were significantly lower than the average UK population [40]. In addition, to the best of our knowledge, this is the first study to report on WPAI activity impairment in thyroid cancer. The WPAI mean activity value in MTC and PTC at data capture was comparable to that reported in locally recurrent or metastatic breast cancer receiving first-line hormonal therapy or chemotherapy and/or targeted therapy [41], suggesting that thyroid cancer has a definite impact on functioning in daily activities similar to other advanced cancers.

Limitations of the study include the relatively small number of patients for whom some variables such as time from recurrence/metastasis to first MKI are available, its retrospective design, and the selection of consecutive patients that may have resulted in over-representation of patients who consult more frequently. Because all the patients included were alive at data capture, evaluation of overall survival was not possible. Also, its point in time design precludes demonstration of cause and effect. In addition, data on the number/type of hospital or cancer center where involved in advanced thyroid cancer treatment were somewhat limited.

Conclusion

This analysis provides real-world evidence about clinical profiles, treatment, RET alteration testing patterns, and HRQoL for patients with MTC and PTC, as well as physician profiles and disease management patterns in Japan. RET testing patterns varied by the type of hospital/center. One of the most commonly stated reasons for not testing was lack of therapeutic relevance. MKIs were the most frequently prescribed systemic therapy but timing to start MKI therapy was variable and adverse events were reported as challenges. The results also suggest that physicians are seeing patients in a wide variety of facilities, with relatively limited opportunities to accumulate experiences to manage thyroid cancer patients. There are also various challenges to overcome to improve rates of RET testing, and potential opportunities to optimize patient outcomes on molecular targeted therapies in order to provide specific, individualized treatment. Our study also highlights that there is substantial burden of both disease and treatment associated with MTC and PTC that leads to deterioration of performance status. Findings suggest a need for more effective and less toxic systemic treatment targeting genomic alterations to improve long-term outcomes for thyroid cancer patients.