Real-world drug utilization and treatment patterns in patients with tenosynovial giant cell tumors in the USA
Abstract
Aim: Real-world treatment patterns in tenosynovial giant cell tumor (TGCT) patients remain unknown. Pexidartinib is the only US FDA-approved treatment for TGCT associated with severe morbidity or functional limitations and not amenable to improvement with surgery. Objective: To characterize drug utilization and treatment patterns in TGCT patients. Methods: In a retrospective observational study using IQVIA's linked prescription and medical claims databases (2018–2021), TGCT patients were stratified by their earliest systemic therapy claim (pexidartinib [N = 82] or non-FDA-approved systemic therapy [N = 263]). Results: TGCT patients treated with pexidartinib versus non-FDA-approved systemic therapies were predominantly female (61 vs 50.6%) and their median age was 47 and 54 years, respectively. Pexidartinib-treated patients had the highest 12-month probability of remaining on treatment (54%); 34.1% of pexidartinib users had dose reduction after their first claim. Conclusion: This study provides new insights into the unmet need, utilization and treatment patterns of systemic therapies for the treatment of TGCT patients.
Plain language summary – Treatment patterns in patients with tenosynovial giant cell tumors in the USA
This database study is the first investigation of how drugs are used to treat patients with tenosynovial giant cell tumor (TGCT) in the real world. We researched adult TGCT patients from IQVIA's prescription and medical claims databases who started treatment with pexidartinib (N = 82) or other non-US FDA-approved systemic therapies (N = 263). The patients included in this analysis were mostly women (61.0 and 50.6%) and their median age was 47 and 54 years for pexidartinib and other non-FDA-approved systemic therapies, respectively. The patients treated with pexidartinib were most likely to remain on treatment (54.0%) at the end of the first year. Most patients (79.3%) started pexidartinib treatment at a total daily dose of 800 mg/day, as per the product label. Only 34.1% of patients had reduced medication dose during follow-up. Of note, this study found that TGCT patients were treated with other systemic therapies which remain unproven to be safe and effective in medical studies of TGCT. Given the unmet need, and with pexidartinib being the only approved systemic treatment in USA, there is an opportunity for the larger population of adult TGCT patients to benefit from its use. Further research is needed to identify barriers for access to pexidartinib and treatment of TGCT patients.
Tenosynovial giant cell tumors (TGCT), formerly referred to as pigmented villonodular synovitis (PVNS), are rare, benign [1] and locally aggressive neoplasms affecting the synovial membrane lining joints, bursae and tendon sheaths [2,3]. Disease presentation is heterogeneous and ranges from mildly asymptomatic to debilitating symptoms, including severe pain, stiffness, swelling and limited range of motion [4]. TGCT presents either as a single nodule, called localized TGCT (L-TGCT) or as multiple nodules, known as diffuse TGCT (D-TGCT) [1,5–10]. From a histopathological and genetic standpoint, both forms are very similar [10,11]; however, their natural course and clinical picture differ significantly [11].
L-TGCT (also known as localized PVNS, giant cell tumor of the tendon sheath [GCT-TS] or nodular synovitis) typically presents as small lobulated lesions roughly 0.5–4 cm in size [10,12]. It is further distinguished based on the location of the tumor, either in digits or larger joints [10,13,14]. Most lesions are found in the digits of the hands and feet [10], with rarer instances in the wrist, ankle, knee and very infrequently in the hip or elbow [13,15]. Around 85% of TGCT cases in the digits are in the localized form, with the majority of lesions arising from the tendon sheath [15]. D-TGCT (diffuse PVNS or synovitis villonodularis pigmentosa [SVP]) is a more destructive and locally aggressive tumor than L-TGCT [10]. Typically, it has a multinodular, multi-colored appearance [12] and arises from the synovial membrane, often affecting adjacent joints or other surrounding structures [10]. Around 75% of diffuse cases are around the knee joint [16], with other cases arising from the hip, ankle or shoulder [10].
In the last 8 years, tyrosine kinase inhibitors (TKIs) that target the colony-stimulating factor 1 receptor (CSF1R) have been used to treat TGCT where surgery is not an option [16–23]. Pexidartinib, a first-in-class selective TKI against CSF1, was approved in August 2019 by the US FDA for treating adult patients with symptomatic TGCT associated with severe morbidity or functional limitations and not amenable to improvement with surgery in the USA [24,25]. Other systemic therapies have been investigated in various studies; although they are approved for multiple other indications, none are approved by the FDA for the treatment of TGCT, specifically. These include therapies targeting the CSF1/CSF1R axis (e.g., imatinib, nilotinib, sorafenib, sunitinib), anti-cytokine biologics (e.g., adalimumab, etanercept) [26] and the anti-receptor activator of nuclear factor-κB ligand (RANKL) antibody denosumab [27]. The National Comprehensive Cancer Network (NCCN) guidelines (2022) currently include only three systemic therapies for the treatment of TGCT: pexidartinib, which has a category 1 preferred recommendation, and imatinib and nilotinib (category 2a), which are useful in certain circumstances. However, NCCN does not provide guidance for non-FDA approved therapies for TGCT [28].
It is assumed that patients with symptomatic TGCT associated with severe morbidity or functional limitations and not amenable to improvement with surgery are being treated with pexidartinib; however, no previous study has assessed this. Before pexidartinib approval, patients were possibly being treated with various systemic therapies not approved for the treatment of TGCT. As pexidartinib is approved by the FDA for treatment of a select group of patients with D-TGCT, not amenable to surgery, there is a significant number of TGCT patients, with varying disease states, otherwise being treated with different nonapproved systemic therapies.
The current pattern of drug utilization among TGCT patients in the real-world setting, in resectable and unresectable patients, is not known in the US population. Although multiple systemic therapies have been investigated and used for the treatment of TGCT, there have not yet been any real-world studies characterizing their associated treatment patterns by treatment regimen. Therefore, this study used large administrative medical and pharmacy claims databases to describe TGCT patients newly treated with available targeted systemic treatments in terms of patient characteristics and treatment patterns, including treatment dosing and duration.
Methods
Study design
This was a retrospective observational study leveraging IQVIA's prescription claims database (LRx) linked to its medical claims database (Dx). Data from 1 February 2018 to 31 March 2021 was used to describe patients newly initiating systemic therapies on or after the FDA approval of pexidartinib (2 August 2019).
Data source
The study dataset was created based on Health Insurance Portability and Accountability Act (HIPAA)-compliant linking processes [29–31]. LRx contains information on dispensed prescriptions with 92% coverage of prescriptions from the retail channel, 72% coverage of standard mail service and 76% coverage of long-term care facilities in the USA. Dx captures over 1 billion pre-adjudicated claims and 3 billion records obtained annually from approximately 800,000 office-based physicians and specialists, with 75% of American Medical Association providers being captured. It also includes medical claims from ambulatory and general healthcare sites, as well as outpatient clinics associated with hospitals such as rehabilitation, same-day surgery and chemotherapy centers. All data were HIPAA-compliant to protect patient privacy. This data source was also utilized in a recent study evaluating the real-world treatment patterns in metastatic non-small-cell lung cancer patients [32]. As this retrospective cohort analysis was conducted using de-identified HIPAA-compliant data, Institutional Review Board (IRB) review was not required for this study.
Patient selection & study cohorts
Inclusion and exclusion criteria for patient selection are described in Figure 1. Prescription claims in LRx or drug administration claims in Dx were used to identify patients treated with pexidartinib, imatinib, nilotinib, sorafenib or sunitinib, where the earliest record of treatment from 1 August 2018 to 28 February 2021 (patient selection window) was the index date to allow for a minimum 6-month baseline period before index and minimum 1-month follow-up period. These drugs were selected for this analysis as they have been more commonly reported as treatments for TGCT in the published literature, while data on adalimumab, denosumab or etanercept are scarce.
1Other systemic therapies included imatinib, nilotinib, sorafenib, sunitinib, adalimumab, denosumab and etanercept. Patients with adalimumab, denosumab or etanercept were later excluded from the final cohorts, due to the limited literature on the use of these medications for the treatment of TGCT.
2Selection of patients into the pexidartinib cohort was prioritized, so that for patients with pexidartinib, the date of the earliest claim for pexidartinib in the selection window was the index date regardless of the presence of other systemic therapies.
3After applying these exclusion criteria, there were 2960 patients remaining.
4Mis-bridged patient IDs are IQVIA patient IDs known to be false positive links combining together one or more real world patients into a single IQVIA patient ID.
5The other systemic therapy cohort (N = 263) consisted of patients with imatinib (N = 109; 37.3%), sorafenib (N = 98; 37.3%), sunitinib (N = 48; 18.3%) and nilotinib (N = 8; 3.0%) on the index date.
Dx: Medical claims database; LRx: Prescription claims database; TGCT: Tenosynovial giant cell tumor.
To ensure that patients received the treatment for TGCT, patients indexed on a systemic therapy other than pexidartinib (‘other systemic therapy’) were required: to have at least one diagnosis of TGCT (defined as International Classification of Diseases, Tenth Revision, Clinical Modification [ICD-10-CM] C49, C41.9, D48.0, D48.1, D48.9, D21.0-D21.9, M12.2, M65.8-M65.9 or M67.9) during the study period and no diagnoses for any of their corresponding approved indications any time before the index date. For example, if a patient was indexed on imatinib with an associated diagnosis of chronic myeloid leukemia, acute lymphoblastic leukemia, myelodysplastic/myeloproliferative diseases, systemic mastocytosis, hypereosinophilic syndrome/chronic eosinophilic leukemia or gastrointestinal stromal tumors, they were excluded from the cohort. If indexed on nilotinib, patients with chronic myeloid leukemia were excluded (Table 1). Two recent studies have also used these diagnosis codes for identifying TGCT patients in claims databases [33,34]. To ensure precision in identifying TGCT, these codes were supplemented with additional codes observed within the pexidartinib cohort (e.g., C49).
| Index systemic therapy | Exclusion criteria |
|---|---|
| Imatinib | • Chronic myeloid leukemia • Acute lymphoblastic leukemia • Myelodysplastic/myeloproliferative diseases • Systemic mastocytosis • Hypereosinophilic syndrome/chronic eosinophilic leukemia • Gastrointestinal stromal tumors |
| Sorafenib | • Hepatocellular carcinoma • Renal cell carcinoma • Thyroid cancer |
| Sunitinib | • Gastrointestinal stromal tumor • Renal cell carcinoma • Pancreatic neuroendocrine |
| Nilotinib | • Chronic myeloid leukemia |
In the medical claims database, it is not possible to confirm TGCT diagnosis using ICD codes. To create a comprehensive list of TGCT-related diagnosis codes, the authors investigated the codes observed within the pexidartinib cohort that could be used to supplement the lists used in other published TGCT studies [32,33]. The study included the parent code ICD-10 C49 (malignant neoplasm of other connective and soft tissue) in the definition of TGCT because the codes under C49 were observed in 11.0% (n = 9) of the pexidartinib cohort on or before the index date. Given that pexidartinib is the only approved for treatment of TGCT, it was assumed that C49 was relevant and was therefore used to identify TGCT in the inclusion criteria for the other systemic therapy cohort. Since patients in the other systemic cohort were excluded if they had diagnosis codes for the corresponding approved indications of imatinib, nilotinib, sorafenib or sunitinib before the index the date, the authors concluded that the final cohort comprised of patients with TGCT only.
Patients meeting all selection criteria were then grouped into two mutually exclusive cohorts defined by their index treatment (pexidartinib cohort and other systemic therapy cohort). Within the other systemic therapy cohort, patients indexed on or after 2 August 2019 were used for all study measures to allow for similar follow-up duration to the pexidartinib cohort. Select measures were additionally reported among patients indexed on other systemic therapies before 2 August 2019. Distribution of ICD-10-CM codes for the pexidartinib cohort and all patients indexed on other systemic therapies are shown in Supplementary Table 1.
Study measures
Patient demographic information (age, sex, geographic region, payer type, index year) and clinical characteristics from the 6-month baseline period, including the Quan-Charlson Comorbidity Index (CCI) [35] and specific comorbidities of interest, were reported. Prior systemic therapies for TGCT, concomitant medications, evidence of prior surgery and number of surgical episodes were reported using pre-index data within the study period beginning 1 February 2018. Surgeries of interest included amputation, arthrodesis, arthroplasty, arthroscopy and excision. Procedure claims from the same surgery type occurring within a 30-day window were considered as part of the same episode.
Follow-up measures were treatment dosage for index and subsequent medication claims (for imatinib, nilotinib, sorafenib or sunitinib), strength of index treatment and index treatment duration. Total daily dose per claim was calculated using the quantity dispensed multiplied by the strength of drug dispensed fields in LRx, and then divided by the days' supply of the claim.
Treatment discontinuation was defined as a gap in therapy of ≥90 days in the medication supply (i.e., after the days' supply on the last claim for the index treatment was exhausted). This duration was selected to allow for gaps between treatment refills of extended periods so that we do not overestimate treatment discontinuation, a practice commonly followed in the analysis of claims database studies. To identify the last day of days' supply for oral medications, overlaps in the days' supply for consecutive claims were ignored to provide a more conservative estimate of treatment duration. Any treatment combinations or modifications were ignored. If the last day of days' supply exceeded the end of the study period, the last day was capped at the end of the study period. If the time from the end of medication supply to the end of follow-up was less than the allowable gap period, then the patient was not eligible to meet the definition of treatment discontinuation.
Statistical analysis
Kaplan–Meier (KM) analysis was used to estimate median and 95% CI for index treatment duration; patients without evidence of index treatment discontinuation were censored at the end of follow-up or the end of the study period, whichever occurred earliest. All analyses were performed using SAS version 9.4 [36].
Results
Epidemiology
The first nationwide study of TGCT, conducted in The Netherlands, reported the estimated incidence rates (IR) in digits, localized extremity, and D-TGCT as 34, 11 and 5 per million person-years, respectively [10]. Another population-based cohort study based on linking patient data from four nationwide Danish registries (1997–2012) reported IRs for localized and diffuse TGCT as 30.3 and 8.4 per million person-years, respectively [37]. Overall prevalence at the end of 2012 per 100,000 persons was 44.3 (95% CI 42.4–46.3) for L-TGCT and 11.5 (95% CI 10.6–12.6) for D-TGCT. There were more prevalent cases among women for both forms. The highest prevalence was seen in the 60–79 year-old age group, followed by the 40–59 year-old group [37]. No population-based studies have been conducted in the USA that provide the incidence and prevalence of TGCT by age, gender or subtype.
To understand the burden of TGCT in the USA, gender-specific TGCT IRs reported by Mastboom et al. in The Netherlands [10] were extrapolated to the gender-specific US population estimates for 2020 [38]. This resulted in an estimated overall population (all ages) annual incidence of 11,028 new cases of L-TGCT of digits (females = 6663 and males = 4365), 3620 new cases of L-TGCT of extremities (females = 2166 and males = 1455) and 1646 new cases of D-TGCT (females = 999 and males = 647). This corresponds to an overall annual incidence of 16,295 new cases of TGCT in the USA for 2020 (localized = 14,649; diffuse = 1646; all ages). For prevalence of TGCT in the adult US population, gender-specific TGCT prevalence reported by Ehrenstein et al. in Denmark [37] were extrapolated to the gender-specific adult (18+ years) US population estimates for 2020 [18]. This resulted in an adult population prevalence of 113,377 cases of L-TGCT (females = 59,482 and males = 43,895) and 29,511 cases of D-TGCT (females = 15,833 and males = 13,678). This corresponds to an overall estimated prevalence of 142,889 cases of adult TGCT in the USA for 2020.
With the available evidence based on the literature, treatment guidelines and extrapolations, serious gaps in knowledge were identified. A sizable TGCT patient population likely exists within the USA, with an estimated 113,777 cases of L-TGCT and 29,511 cases of D-TGCT for the year 2020. Based on the studies discussed in the sections above, approximately 49% of L-TGCT patients are treated with surgery, and 15% of these patients will require more than one surgery. This means there are approximately 50% of patients with L-TGCT who do not get surgery. Additionally, there is a gap in the literature on the treatments that these patients receive. Similarly, approximately 84% of D-TGCT patients are treated with surgery, and 20–50% of these patients will require multiple surgeries.
Surgery may be curative for some TGCT patients (mainly L-TGCT); it is difficult to completely remove diffuse tumors. A single surgical resection alone may be inadequate to control D-TGCT for many patients. Moreover, many of these tumors recur after resection. Also, following re-resection of recurrent tumors, joint arthroplasty may be required to restore mechanical joint function [3,10,39]. With any surgery, there is a risk of surgical complications, regardless of the type of surgery. Logically, repeated surgeries would increase the risk of surgical complications.
With these factors in mind, a question remains: what treatments are available for those TGCT patients who are not candidates for surgery? Some patients who are watching and waiting will eventually require treatment of some sort as well. A conservative estimate of at least 50% of L-TGCT and 15% of D-TGCT patients with symptomatic or unresectable disease are potential candidates of any drug treatment (systemic or symptomatic relief); this corresponds to 56,527 L-TGCT and 4402 D-TGCT patients in the USA in the year 2020.
Demographic & clinical characteristics
The study included a total of 345 adults treated with a systemic therapy for TGCT (82 patients in the pexidartinib cohort and 263 in the other systemic therapy cohort; Figure 1). The other systemic therapy cohort (N = 263) consisted of patients receiving one of the non-pexidartinib systemic therapies for TGCT: imatinib (N = 109; 37.3%), sorafenib (N = 98; 37.3%), sunitinib (N = 48; 18.3%) and nilotinib (N = 8; 3.0%).
The demographic and baseline clinical characteristics of the pexidartinib and other systemic therapy cohorts are shown in Tables 2 & 3. The median ages of the pexidartinib cohort and the other systemic therapy cohort were 47 and 54 years, respectively. Both cohorts were predominantly female (61.0 and 50.6%, respectively). Of the pexidartinib cohort, 62.2% of patients were 40–64 years old, while 39.9% of patients in the other systemic therapy cohort were 40–64 years old. Only 4.9% of patients were 65 years or older in the pexidartinib cohort versus 31.2% in the other systemic therapy cohort.
| Study measures | Pexidartinib cohort (N = 82) | Other systemic therapy cohort (N = 263) |
|---|---|---|
| Age (years) on index date | ||
| Mean (SD) | 45.7 (12.0) | 52.5 (19.6) |
| Median (Q1, Q3) | 47 (38, 54) | 54 (36, 69) |
| Age group (years) (n, %) | ||
| 18–24 | 3 (3.7) | 24 (9.1) |
| 25–39 | 24 (29.3) | 52 (19.8) |
| 40–54 | 35 (42.7) | 62 (23.6) |
| 55–64 | 16 (19.5) | 43 (16.3) |
| 65–74 | 3 (3.7) | 39 (14.8) |
| 75+ | 1 (1.2) | 43 (16.3) |
| Sex (n, %) | ||
| Female | 50 (61.0) | 133 (50.6) |
| Male | 32 (39.0) | 130 (49.4) |
| Geographic region (n, %) | ||
| Northeast | 25 (30.5) | 56 (21.3) |
| West | 24 (29.3) | 52 (19.8) |
| South | 19 (23.2) | 96 (36.5) |
| Midwest | 14 (17.1) | 59 (22.4) |
| Payer type (n, %) | ||
| Commercial | 75 (91.5) | 185 (70.3) |
| Medicare Part D | 5 (6.1) | 58 (22.1) |
| Medicaid | 2 (2.4) | 7 (2.7) |
| Medicare | 0 (0.0) | 7 (2.7) |
| Cash payments | 0 (0.0) | 6 (2.3) |
| Study measures | Pexidartinib cohort (N = 82) | Other systemic therapy cohort (N = 263) |
|---|---|---|
| Quan-CCI | ||
| Mean (SD) | 0.5 (1.3) | 2.3 (2.5) |
| Median (Q1, Q3) | 0 (0, 0) | 2 (0, 3) |
| CCI category (n, %) | ||
| 0 | 62 (75.6) | 109 (41.4) |
| 1–2 | 14 (17.1) | 68 (25.9) |
| 3+ | 6 (7.3) | 86 (32.7) |
| CCI comorbidities (n, %) | ||
| Any malignancy† | 8 (9.8) | 79 (30.0) |
| Chronic pulmonary disease | 6 (7.3) | 15 (5.7) |
| Diabetes – without chronic complications | 3 (3.7) | 12 (4.6) |
| Diabetes – with chronic complications | 2 (2.4) | 3 (1.1) |
| Liver disease – mild | 2 (2.4) | 21 (8.0) |
| Peripheral vascular disease | 2 (2.4) | 7 (2.7) |
| Congestive heart failure | 1 (1.2) | 6 (2.3) |
| Metastatic solid tumor | 1 (1.2) | 55 (20.9) |
| Myocardial infarction | 1 (1.2) | 0 (0.0) |
| Paraplegia and hemiplegia | 1 (1.2) | 2 (0.8) |
| Renal disease | 1 (1.2) | 11 (4.2) |
| Rheumatologic disease | 0 (0.0) | 1 (0.4) |
| Cerebrovascular disease | 0 (0.0) | 3 (1.1) |
| Peptic ulcer disease | 0 (0.0) | 2 (0.8) |
| Dementia | 0 (0.0) | 1 (0.4) |
| HIV/AIDS | 0 (0.0) | 1 (0.4) |
| Liver disease – moderate to severe | 0 (0.0) | 0 (0.0) |
| Other comorbidities or differential diagnoses of interest (n, %) | ||
| Osteoarthritis | 13 (15.9) | 16 (6.1) |
| Hypertension | 8 (9.8) | 42 (16.0) |
| Obesity | 7 (8.5) | 18 (6.8) |
| Fibroma‡ | 7 (8.5) | 17 (6.5) |
| Connective and soft tissue cancers† | 6 (7.3) | 104 (39.5) |
| Type 2 diabetes | 5 (6.1) | 15 (5.7) |
| Infections | 1 (1.2) | 11 (4.2) |
| Meniscal tear | 1 (1.2) | 1 (0.4) |
| Rheumatoid arthritis | 0 (0.0) | 1 (0.4) |
| Gout | 0 (0.0) | 2 (0.8) |
| Systemic lupus erythematosus | 0 (0.0) | 0 (0.0) |
The majority of patients in the pexidartinib cohort were located in either the north eastern (30.5%) or western (29.3%) USA, and more than one third (36.5%) of the other systemic therapy cohort was located in the southern USA. On the index medication claim, 91.5 and 70.3% of the pexidartinib and other systemic therapy cohorts, respectively, had commercial payers. Only 6.1% of the pexidartinib cohort were on Medicare Part D versus 22.1% of patients in the other systemic therapy cohort. The pexidartinib cohort had a median CCI score of 0, whereas the other systemic therapy cohort had a median CCI score of 2, indicating patients in the other systemic therapy cohort were more likely to have comorbid conditions than patients in the pexidartinib cohort. The most common comorbidity was osteoarthritis (15.9%) for the pexidartinib cohort and connective and soft-tissue cancers (39.5%) for the other systemic therapy cohort.
When looking at patients indexed 1 August 2018 to 1 August 2019 (pre-pexidartinib approval), an additional 199 patients were indexed on other systemic therapies. The baseline clinical and demographic information was nearly identical to the other systemic therapy cohort post-pexidartinib approval (Supplementary Table 2). The median age was 53 years and 55.8% were female. The age distribution as well as CCI scores were nearly identical to the other systemic therapy cohort post-pexidartinib approval.
Treatment history
Treatment history was described using all available pre-index data. The median duration (Q1, Q3) of the pre-index period was 25 (22, 31) months for the pexidartinib cohort and 25 (21, 30) months for the other systemic therapy cohort. Evidence of at least one TGCT-related outpatient surgery was seen in 15.9% of the pexidartinib cohort and 9.5% of the other systemic therapy cohort (Table 4). This low percentage is due to the short duration of the study period. Among patients in the pexidartinib cohort, the most common surgery types were excision (13.4%) and arthroscopy (3.7%); in the other systemic therapy cohort, 4.6% of patients had excision and 1.1% had arthroscopy. When outpatient surgery using all available pre-index data since January 2010 was analyzed, 37.8% of the pexidartinib cohort and 15.6% of the other systemic therapy cohort had evidence of at least one TGCT-related outpatient surgery. Among patients in the pexidartinib cohort, the most common surgery types were excision (28.0%) and arthroscopy (17.1%); in the other systemic therapy cohort, 6.1% had excision and 4.2% had arthroscopy.
| Treatment history | Pexidartinib cohort (N = 82) | Other systemic therapy cohort (N = 263) |
|---|---|---|
| Outpatient surgery (n, %) | ||
| Any surgery | 13 (15.9) | 25 (9.5) |
| Excision | 11 (13.4) | 12 (4.6) |
| Arthroscopy | 3 (3.7) | 3 (1.1) |
| Arthroplasty | 2 (2.4) | 7 (2.7) |
| Arthrodesis | 0 (0.0) | 2 (0.8) |
| Amputation | 0 (0.0) | 5 (1.9) |
| Number of surgical episodes (n, %) | ||
| 0 | 69 (84.1) | 238 (90.5) |
| 1 | 10 (12.2) | 20 (7.6) |
| 2 | 3 (3.7) | 3 (1.1) |
| 3+ | 0 (0.0) | 2 (0.8) |
| Systemic therapies for TGCT‡ (n, %) | ||
| Any of the below | 8 (9.8) | 5 (1.9)¶ |
| Imatinib | 8 (9.8) | 3 (1.1) |
| Sorafenib | 0 (0.0) | 0 (0.0) |
| Sunitinib | 0 (0.0) | 3 (1.1) |
| Nilotinib | 0 (0.0) | 0 (0.0) |
| Concomitant medications§ (n, %) | ||
| Any of the below | 53 (64.6) | 193 (73.4) |
| Antibiotics | 28 (34.1) | 120 (45.6) |
| Opioids | 26 (31.7) | 131 (49.8) |
| NSAIDs | 19 (23.2) | 62 (23.6) |
| Systemic corticosteroids | 11 (13.4) | 50 (19.0) |
| Non-narcotic analgesics | 2 (2.4) | 2 (0.8) |
| Topical anti-inflammatories | 2 (2.4) | 8 (3.0) |
| DMARDs | 0 (0.0) | 10 (3.8) |
| Antirheumatics | 0 (0.0) | 0 (0.0) |
Evidence of a systemic therapy for TGCT other than the index therapy was infrequent and was observed in 9.8% of the pexidartinib cohort, all with pre-index imatinib, and 1.9% of the other systemic therapy cohort. The majority of patients (64.6% of the pexidartinib cohort and 73.4% of the other systemic therapy cohort) had received at least one concomitant medication of interest, most commonly antibiotics (34.1%), opioids (31.7%) and nonsteroidal anti-inflammatory drugs (NSAIDs; 23.2%) in the pexidartinib cohort and opioids (49.8%), antibiotics (45.6%) and NSAIDs (23.6%) in the other systemic therapy cohort (Table 4).
Patterns of ICD-10-CM codes for TGCT differed between the pexidartinib cohort and the other systemic therapy cohort indexed at any time, with patients in the pexidartinib cohort having higher frequencies of certain diagnosis codes, notably M12.2 (villonodular synovitis, pigmented), D21.2 (benign neoplasm of connective/soft tissue), and M65.9 (synovitis and tenosynovitis, unspecified), whereas the other systemic therapy cohort was generally more likely to have diagnosis codes for C49.9 (malignant neoplasm of connective and soft tissue, unspecified). The comparison of distributions of diagnosis codes observed before or on the index date for both cohorts are described in detail in Supplementary Table 1.
Dosage patterns of systemic therapies for TGCT
The median length (Q1, Q3) of follow-up was 11.8 (7.1, 15.6) months for the pexidartinib cohort and 10.3 (6.7, 15.4) months for the other systemic therapy cohort for describing treatment patterns. Total daily dose prescribed or administered based on each claim for TGCT treatments of interest was generally consistent with the product label-recommended dose for their respective approved indications. For example, when considering all prescription claims by treatment, 79.3% of pexidartinib-treated patients had claims with a total daily dose of 800 mg/day (label recommendation for TGCT), 39.0% had claims with a total daily dose of 400 mg/day and 26.8% had claims with a total dose of 600 mg/day (both are label recommendations for dose reduction).
The most common total daily dosages were 400 mg/day (84.4%) and 800 mg/day (9.2%) for imatinib-treated patients, 400 mg/day (73.5%) and 800 mg/day (27.6%) for sorafenib, 37.5 mg/day (66.7%) and 25 mg/day (16.7%) for sunitinib, and 800 mg/day (75.0%) and 600 mg/day (25.0%) for nilotinib. Furthermore, for the index (i.e., first) oral medication claim, the mean total daily dosage was 713.4 mg/day for pexidartinib, 411.9 mg/day for imatinib, 492.9 mg/day for sorafenib, 36.2 mg/day for sunitinib and 743.3 mg/day for nilotinib.
Across all observed claims during follow-up, the mean dosage was numerically lower compared with claims on the index date (Figure 2). For example, the mean dosage of all pexidartinib claims was 663.6 mg/day. In addition, 0–34.1% of patients had a lower dosage on their subsequent claim compared with the index claim, including 34.1% (N = 28) of the pexidartinib-treated patients. A higher dosage on a subsequent claim compared with the index claim was infrequently observed, ranging from 5.5 to 12.5%, including 9.8% (N = 8) of the pexidartinib-treated patients (Figure 3).
The mean total daily dose is calculated based on the number of claims corresponding to each measure. For the ‘Index claim’ category, the number of claims corresponds to 82, 109, 98, 48 and 8 for pexidartinib, imatinib, sorafenib, sunitinib and nilotinib, respectively (consistent with the patients with the index treatment). For the ‘Any claim’ category, the number of claims corresponds to 556, 583, 465, 170 and 33. For the ‘Last claim’ category, the number of claims corresponds to 33, 60, 56, 31 and 7 (consistent with the patients who discontinued the index treatment).
The denominators for the calculations are the number of patients with the corresponding index medication.
Dosage was further described among patients who discontinued oral medications. The mean total daily dosage of the last observed claim before treatment discontinuation was numerically greater than the dosages among all patients any time during follow-up (Figure 2). For example, the mean total daily dosage for any pexidartinib claim was 663.6 mg/day, while patients who discontinued pexidartinib had a mean dosage of 690.9 mg/day before they discontinued treatment.
Duration & discontinuation of the index treatment
The median treatment durations varied across the TGCT therapies, ranging from 2.2 to 5.1 months (Figure 4 & Table 5). The median duration of pexidartinib treatment was not reached over the follow-up period. The probability of remaining on treatment at 12 months after treatment initiation also varied across the treatment regimens, ranging from 17.1% for sunitinib to 54.0% for pexidartinib (Table 5).
| Treatment | Patients who discontinued treatment anytime over follow-up n (%) | Probability of remaining on treatment at 12 months (%) | Treatment duration, median months (95% CI) |
|---|---|---|---|
| Pexidartinib (N = 82) | 33 (40.2) | 54.0 | NE (7.1, NE) |
| Imatinib (N = 109) | 60 (55.0) | 36.4 | 5.1 (2.8, 11.2) |
| Sorafenib (N = 98) | 56 (57.1) | 31.7 | 4.4 (2.7, 8.5) |
| Sunitinib (N = 48) | 32 (66.7) | 17.1 | 3.6 (1.9, 5.4) |
| Nilotinib (N = 8) | 7 (87.5) | NE | 2.2 (0.9, 6.3) |
A small portion of patients who met the definition of index treatment discontinuation in our study had evidence of restarting the treatment they discontinued; this ranged from 6.3 to 21.2% of patients who discontinued treatment (Figure 5). For example, of the 33 patients whom we considered to have discontinued pexidartinib, 21.2% (N = 7) restarted pexidartinib after a gap of 90 days or more in medication supply.
Discussion
For L-TGCT, the standard of care is marginal excision [1,10,40]. A systematic literature review reported that after an average follow-up of 37 to 79 months, L-TGCT patients had a recurrence of 15% regardless of the type of surgery [41]. After arthroscopic and open resection, an average recurrence of 6 and 4% has been reported, respectively [10,42]. However, 73–91% of patients have shown no local recurrence at 5 years in the most recent studies [5,15,43]. For D-TGCT, the standard of care is still surgical excision [1,39,44]. Average recurrence rates of 40 and 14% have been reported after arthroscopy and open resection, respectively [42]. However, 27–62% of patients with D-TGCT have been reported to experience no local recurrence at 5 years [15]. An international retrospective cohort study assessed surgical outcomes of patients with D-TGCT with 282 recurrent cases and 910 primary cases. The main affected sites included the knee (63.6%), hip (10.4%), ankle (13.6%) and foot (5.3%). Types of surgeries performed were arthroscopic synovectomy (13.7%), one-staged open synovectomy (54%), two-staged open synovectomy (16.1%), prosthesis (4.6%) and amputation (0.3%). At a median follow-up of 54 months, recurrent disease had developed in 44% of the 966 surgically treated cases [45].
Burton et al. analyzed claims data for commercial and Medicare Advantage health plan enrollees in the USA (n = 4664) using diagnosis codes for GCT-TS (ICD-9: 727.02) and PVNS (ICD-9: 719.2x). Approximately 50% of patients had surgery during the 12-month post-index period; 84.2% of GCT-TS subjects were treated with surgery and 15.5% received no treatment. Among PVNS subjects, 48.7% had surgery and 50.7% had no specified post-index treatment [33]. Another retrospective claims analysis (2011–2015) included adult TGCT patients (diagnosed using ICD-9 codes: 727.02, 719.2x; ICD-10: D48.1, D21.x, M12.2); 13,994 TGCT patients were followed for a median of 28.5 months. In the post-index period, 7199 patients had ≥1 surgery, with 73.0, 21.0 and 3.4% undergoing 1, 2 and ≥3 surgeries, respectively. The median time from diagnosis to the first surgery was 563 days. The 1- and 2-year risk of recurrence after the first surgery was 21.51 and 29.17%, respectively. For those that received two surgeries, the 1- and 2-year risk of recurrence after the second surgery was 18 and 29%, respectively [46]. Lin et al. evaluated newly diagnosed TGCT patients (18–64 years) included in the OptumHealth Care Solutions, Inc. database who had joint surgery during 1999 and 2017. Of the 835 TGCT patients included in this study, 55% who had ≥1 joint surgery were followed for a median of 5.7 years. During year 1, 78% patients had their first joint surgery and 41% had more than one repeat surgery [39].
An observational, multinational registry (2016–2019) including 166 TGCT patients reported the use of systemic treatments in 52 patients, with the majority of patients enrolled in the European Union [23]. The most common treatments were pexidartinib and imatinib, mostly in patients with locally advanced refractory disease that may not be improved by surgery. Treatment strategies at baseline were watchful waiting (48.8%), surgery (24.7%), or targeted systemic therapy (22.3%). The mean age of patients was 44 years, 83.7% received prior treatments; 42.8% had ≥1 recurrence after treatment of their primary tumor. Prior treatment consisted of surgery (77.1%) and systemic treatments (31.3%) with imatinib (36.5%), pexidartinib (51.9%) and unspecified systemic therapies (11.5%) [23].
To our knowledge, this is the first real-world study in the USA leveraging a large administrative claims database to describe characteristics and treatment patterns of patients newly treated with pexidartinib as well as patients treated with other non-FDA-approved targeted systemic therapies that have been investigated for the treatment of TGCT. Our study included TGCT patients treated with pexidartinib (N = 82) and other systemic therapies, including imatinib (N = 109), sorafenib (N = 98), sunitinib (N = 48) and nilotinib (N = 8). Patients treated with adalimumab, denosumab and etanercept were excluded from this study as these drugs have not been studied in trials and non-interventional studies for TGCT. The efficacy and safety of these drugs has not been demonstrated in controlled clinical trials for treatment of TGCT.
For the pexidartinib cohort, patient demographics, baseline clinical characteristics and treatment history in our study were consistent with prior studies on TGCT [33,34,47,48]. The median age of the pexidartinib cohort in our study was 47 years, which was consistent with previously published cohort studies of real-world TGCT patients (ranging from 41 to 51 years old) [33,34,47,48]; however, patients in the other systemic therapy cohort had a median age of 54 years, older than that in previously published studies of TGCT patients. Given what is known about TGCT, localized and diffuse, the indication for pexidartinib as well as the significant differences in specific TGCT diagnosis codes for the two cohorts, the logical conclusion is that those treated with pexidartinib would have symptomatic TGCT not amenable to surgery while those treated with other non-FDA-approved systemic therapies could plausibly be a different subset of TGCT. Further, since the optimal duration of pexidartinib treatment in the real world setting is not known, it is necessary to identify patients who would derive maximal benefit by evaluating their disease progression and their relative morbidity risks for surgery and other treatment options [49]. The female predominance of our study cohorts is also anticipated for a TGCT population [33,34,47,48], as well as a low baseline comorbidity burden [33,34,45]. As anticipated, patients in the other systemic therapy cohort appeared to have a higher comorbidity burden than the pexidartinib cohort (mean CCI of 2.3 vs 0.5) and a numerically higher proportion with Medicare Part D plans (22.1 vs 6.1%). This is likely due to their advanced age. In addition, congruent with pexidartinib label recommendation [8], evidence of mild liver disease (N = 2) was rarely observed in the pexidartinib cohort and there were no patients with severe liver disease (Table 3).
Indexed patients who were evaluated prior to the approval of pexidartinib, 1 August 2018 to 1 August 2019, reveal a similar cohort of other systemic therapy patients who were indexed post-pexidartinib approval (Supplementary Table 2). This cohort is nearly identical in every way, from demographics to previous treatment history, indicating that there was a consistent population of possible TGCT patients being treated with systemic therapies prior to pexidartinib approval. If the study window was extended, more patients would likely show up in the database. This highlights a gap in care for TGCT patients, who are being prescribed non-FDA-approved systemic therapies that have not demonstrated efficacy and safety in the clinical trial setting within the TGCT population.
In terms of treatment history for TGCT, using all available pre-index data since 2010 (median of 109 months for the pexidartinib cohort and 107 months for the other systemic therapy cohort), evidence of a TGCT-related surgery prior to systemic therapy treatment was observed in 37.8% of the pexidartinib cohort and 15.6% of the other systemic therapy cohort. This post-hoc analysis added an additional 6–20% more surgeries captured compared using data within the study period beginning 1 February 2018. Frequent use of analgesics was anticipated to reduce pain as TGCT presents with joint destruction, inflammation, pain or tenderness, swelling and stiffness [6]. A previous real-world cohort study of TGCT patients reported opioid use in more than 60% of patients during a 12-month follow-up period following the index TGCT diagnosis [44]. This was about a 10% increase in utilization from the pre-index period. Our study found less use of opioids across both the pexidartinib (31.7%) and other systemic therapy cohort (49.8%) during the pre-index period. Post-index utilization of concomitant medications was not evaluated in this study.
It is expected that clinical trial populations differ from patients treated in the real-world setting [50,51]. The most commonly recorded dosages of all systemic therapies, not approved by the FDA for treatment of TGCT, were consistent with the product labels for their respective approved indications [52–55]. The mean total daily dosage for the initial pexidartinib claim was 713 mg/day, which suggests that most physicians follow the recommended starting dose; however, some physicians may start patients on lower dosages to assess tolerability to treatment. Dose reduction was observed in about one-third of the pexidartinib cohort (N = 28; 34.1%). Patients on pexidartinib had a mean dosage of 690 mg/day before they discontinued treatment. For those patients that discontinued pexidartinib due to intolerance, it could be attributed to lack of dose reduction. The pexidartinib label recommends a dose reduction to 600 mg/day for tolerance issues and a further reduction to 400 mg/day is possible. Patients discontinuing on an average of 690 mg/day suggests that the proper dose reduction strategy was not being utilized. It is likely that those that take advantage of dose reductions have a greater ability to stay on this drug longer. Dose reduction among patients treated with other systemic therapies was infrequent (N = 28 across all non-pexidartinib therapies, including 14.6% of patients with sunitinib, 11.2% of patients with sorafenib and 9.2% of patients with imatinib). Determining why these dose reductions occurred within the other systemic therapy cohort is not possible. Any number of reasons could be at play, including adverse events and tolerability. Some patients actually had dose increases following the index claim of pexidartinib (N = 8) and other oral systemic therapies (N = 21), suggesting a cautious approach to assess tolerability. Of the 40.2% of the pexidartinib cohort who met the definition of index treatment discontinuation in our study, 21.2% had evidence of restarting pexidartinib, which might reflect treatment gaps occurring during pexidartinib treatment (Figure 5).
In our study, the median index treatment duration of pexidartinib was not estimable during the follow-up period (mean 11.8 months). This was anticipated because the median treatment duration of pexidartinib in the ENLIVEN trial was 19 months over 26 months of follow-up [20]. Existing literature on treatment duration of nilotinib and imatinib, based on a retrospective multi-institutional chart review study of 62 imatinib-treated patients [17] and a retrospective single-institution chart review study of 15 imatinib-treated patients and four nilotinib-treated patients [56], reported median treatment durations that are consistently a few to several months longer than what our study observed. This is likely driven by the longer follow-up lengths in the previous studies compared with our limited follow-up duration. In addition, the estimated probability of remaining on the index treatment at 12 months varied across the treatments, with the highest probabilities observed for pexidartinib users (54.0%), followed by patients indexed on imatinib (36.4%) and sorafenib (31.7%).
Palmerini et al. conducted an international, observational, prospective study that reported on treatment strategies and outcomes of patients with diffuse TGCT. Of the 176 patients included in that study, 47 (26.7%) either received or planned to receive surgery, 48 (27.4%) received or planned to receive systemic therapy (44/48 at baseline, 4/48 within 2 years), and 6 (3.4%) received or planned to receive radiotherapy (including 90 Yttrium therapy). Of those 44 patients undergoing systemic treatment at baseline, 37 (84.1%) received systemic therapy (i.e., pexidartinib, imatinib, nilotinib, investigator study medicine) while the other 7 (15.9%) were treated with surgery. Further treatment was planned for 28.4% patients. Planned treatment types for those patients were surgery (72%, n = 36), systemic therapy (22.0%, n = 11) and radiotherapy (12%, n = 6). For patients that had systemic therapy at baseline, median pain interference, pain severity, worst pain and worst stiffness measures were numerically lower for those that stayed on systemic therapy versus those that switched therapies at year one. Additionally, for those that remained on systemic therapy for 2 years, scores for the same measures improved. Palmerini et al. provide additional context to this study as in diffuse TGCT alone, there is systemic treatment variation and systemic therapies improve symptomatic disease according to patients [57].
The NCCN guidelines lists three systemic treatments for TGCT: pexidartinib (preferred), imatinib (useful under certain circumstances) and nilotinib (useful under certain circumstances). However, it is not known what these ‘certain circumstances’ are. The only other information available from the NCCN guidelines are references to three studies evaluating their efficacy, one for each of the three systemic treatments [28]. Pexidartinib, the preferred treatment, was evaluated in the two-part phase III ENLIVEN trial against placebo in adult symptomatic advanced TGCT patients for whom surgery was not recommended. The efficacy of pexidartinib was assessed using Response Evaluation Criteria in Solid Tumors (RECIST) at 25 weeks. The overall response (complete or partial) for patients on pexidartinib was 39%, only based on the double-blind portion of the trial. Overall response increased to 53% when the open-label portion of the trial was included [20].
A retrospective study, conducted at 12 referral centers in Europe, USA and Australia, evaluated all consecutive TGCT patients treated with imatinib. There were no specific inclusion/exclusion criteria mentioned in the study or mention of how patients were selected to receive imatinib. There were 29 patients that were included in the study with 69% of them having undergone previous surgery. Metastatic relapse, confirmed by biopsy, occurred in two patients after two operations for local disease. It was a malignant transformation of TGCT/PVNS, confirmed histologically. Efficacy of imatinib was measured by using RECIST best response for 27 patients. Two study subjects were not evaluable due to cessation of their treatment with imatinib mesylate. The first stopped treatment because of neutropenic fever, and the second due to the development of grade 3 edema prior to first radiologic evaluation. Of the 27 patients evaluated using RECIST, only one patient had complete response and four patients had partial response [23]. Nilotinib was evaluated in a multicenter, open-label, single-arm, phase II study. Patients were eligible if they had inoperable progressive or relapsing PVNS or were only resectable with mutilating surgery. Patients were evaluated at 12 weeks using RECIST for the primary end point of progression-free survival (meaning stable disease using RECIST or better). Of the 56 patients originally enrolled, 51 were evaluated. At week 12, 5 (7.6%) of the 56 patients were not evaluable. At week 12, none (0%) of the patients had an objective response. After 1 year of treatment, 46 (90%) of 51 patients achieved the best overall response as stable disease, and 3 (6%) of 51 patients achieved a partial response. Of patients receiving a partial response, two partial responses were observed at 24 weeks and one was observed at 1 year. An objective response was achieved in 6% patients during the 1-year study period. Treatment discontinuation occurred in 25 (45%) of 56 patients before 1 year. Of these 25 patients, treatment failure occurred in 22 (39%) patients [22]. Other systemic treatments that were included in the current study, sorafenib and sunitinib, were not listed as potential treatment options under the NCCN guidelines; however, their use has been documented in published literature, though efficacy and safety within the TGCT population has not been established. This analysis showed a significant proportion of TGCT patients being treated with sorafenib (N = 98; 37.3% of other systemic therapy cohort) and sunitinib (N = 48; 18.3%), where efficacy and safety are largely unknown. As for imatinib and nilotinib, 109 and eight patients were treated using these agents in our study, respectively.
Given the significant TGCT population identified through this study, it is evident that there are different treatment strategies involved in the general treatment of TGCT, whether localized or diffuse. Systemic treatments are used, regardless of their approval for the specific indication. There is an obvious unmet need in patients and these systemic treatments are being utilized as a stopgap measure to improve symptoms in TGCT patients in the real-world setting. However, the use of systemic treatments may differ by region; in a randomized, double-blind clinical trial conducted by Takeuchi et al. in Japan, none of the 41 enrolled patients with D-TGCT or unresectable L-TGCT in the knee joint or ankle had previous use of systemic therapy or radiotherapy [58]. In addition, variability in treatment of TGCT depending on the setting where the patients are diagnosed is likely. Mastboom et al. highlighted that there is no centralization of care for TGCT patients; only 12 and 18% of TGCT and D-TGCT patients were primarily treated in a tertiary oncology center, respectively [4]. Although this study was conducted in The Netherlands [4], similar diagnosis and referral pattern can be expected in other Western countries.
Limitations
There are several limitations of this study inherent to retrospective studies using administrative claims databases, including lack of clinical data to confirm TGCT diagnosis identified using ICD codes, resulting in a possibility that not all patients identified in this study had TGCT. To address this limitation, we required evidence of a TGCT diagnosis code for all patients in our study, and since pexidartinib is currently only indicated for the treatment of severe TGCT that is not amenable to surgery, it is expected that all patients included in the pexidartinib cohort did indeed have TGCT. We also excluded patients from the other systemic therapy cohort with diagnosis codes for their respective non-TGCT treatment indications to help ensure that the index treatment was for TGCT. Furthermore, LRx and Dx are limited to capture of outpatient healthcare resource utilization. Treatments received in the inpatient setting would not be captured in this study. In addition, our study lacked clinical data that would provide insights into reasons for treatment discontinuation, dosage decrease and treatment restarts during follow-up, as well as data on type of TGCT (i.e., localized or diffuse), location of disease and treatment-related adverse events. The limited available duration of pre-index data for capturing surgeries via procedure codes likely resulted in an underestimate of surgical history, as well. To address this, we conducted a post-hoc analysis using all available pre-index data from January 2010, which resulted in a larger number of surgeries being captured.
Last, the study databases were open-source databases where continuous enrollment cannot be confirmed; while proxies for stable enrollment were applied in patient selection, these do not guarantee continuous observation of patients during the study period. Hence, treatment history and index treatment persistence may have been underestimated if patients had additional claims from pharmacies or medical providers that did not contribute data to the LRx or Dx databases.
This study provides novel findings in a population that has not yet been well-characterized using real-world data in the USA. The study leveraged the LRx database which has industry-leading coverage of prescription claims data; only a few essential patient exclusion criteria were applied to minimize patient attrition. As a result, our study is expected to capture most patients treated with pexidartinib since its FDA approval, and the results are likely generalizable to other TGCT patients treated with pexidartinib during the study period in the real-world setting.
Conclusion
This study highlights the significant unmet need for treating patients with TGCT. In addition, it provides insights into the real-world profiles of TGCT patients and their treatment patterns. Results from this study show that D-TGCT patients not amenable to surgery are being treated with pexidartinib, as per the label, and a separate subgroup of TGCT patients is being treated with non-FDA-approved systemic therapies. This is evident by the differences in patient characteristics and TGCT diagnosis codes for these subgroups. TGCT patients treated with non-FDA-approved systemic therapies' demographics and drug utilization patterns were similar to pre-and post-Pexidartinib approval. This suggests that this subset of patients may have a different disease profile, and further research is critical to understand their unique characteristics and establish the optimal treatment options.
Results from this study show that at 12 months, the majority (54%) of patients treated with pexidartinib remained on treatment. The dose of pexidartinib in the real-world setting was generally consistent with dose levels in the pexidartinib label, with dose reduction observed in only one-third of the patients. The dose levels of the different non-FDA-approved systemic therapies used were consistent with their recommended dosages for other indications, as per their label, none of which were TGCT, and 63.6–82.9% of these TGCT patients discontinued treatment at 12 months.
Since its approval, only 0.4% of the estimated 142,889 prevalent patients with TGCT in the USA have been treated with pexidartinib. One of the possible explanations for the lower-than-expected utilization of pexidartinib could be due to healthcare professionals' perceptions of the Pexidartnib Risk Evaluation and Mitigation Safety (REMS) program. Although the Pexidartinib REMS assessment results have shown it to mitigate the potential risk of hepatoxicity successfully, limited use of pexidartinib suggests that the REMS program may be perceived as a barrier to its access.
Given the large unmet need for treating TGCT patients, and as pexidartinib is the only regulatory-approved systemic therapy to date, further research is needed to identify barriers to access to pexidartinib and its utilization.
Tenosynovial giant cell tumor (TGCT) is a rare, locally aggressive neoplasm affecting the synovial membrane lining joints, bursae and tendon sheaths.
Pexidartinib is currently the only systemic treatment approved by the US FDA for the treatment of adult patients with symptomatic TGCT associated with severe morbidity or functional limitations and not amenable to improvement with surgery. Other systemic therapies have been investigated in clinical studies of TGCT patients, none of them have been approved by the FDA for the treatment of TGCT specifically.
Using a real-world claims database, this study characterized TGCT patients newly treated with pexidartinib or other unapproved systemic treatments and described their treatment patterns, including treatment duration and dosing.
In this study, we found that although the median duration of pexidartinib treatment was not reached, patients on pexidartinib had the highest 12-month probability of remaining on treatment.
Dose reductions after the first medication claim were commonly observed in the pexidartinib cohort (34.1%) and some patients (9.8%) had evidence of dose increase.
Dose levels of the non-FDA-approved systemic therapies were consistent with the recommended dosages for other indications, as per their label, none of which were TGCT. As expected, 63.6–82.9% of TGCT patients treated with non-FDA-approved systemic therapies discontinued treatment at 12 months.
This study provided novel insights into how systemic therapies for TGCT are used in real-world clinical practice, where surgery has been the standard of care.
Given the large unmet need for the treatment of TGCT patients, and as pexidartinib is the only regulatory-approved systemic treatment, there is opportunity for its wider use in this patient population.
Supplementary data
To view the supplementary data that accompany this paper please visit the journal website at: www.futuremedicine.com/doi/suppl/10.2217/fon-2023-0363
Author contributions
All authors provided substantial and equal contributions (A) to the conception and design of this research; for the analysis and interpretation of data; and (B) toward drafting the work and revising it critically for important intellectual content; and (C) for final approval of the version to be published; and (D) are in agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Financial disclosure
C Dharmani, E Wang, M Salas, M Wooddell and N Tu are full-time employees of Daiichi Sankyo. All own restricted stock units of Daiichi Sankyo (except E Wang). O Fofah is a postdoctoral fellow employed by Rutgers University and has no financial disclosures to report. A Near and J Tse are full-time employees of IQVIA. G Tinoco is a full-time employee of The Ohio State University Wexner Medical Center. He served on an Advisory Board with Daiichi Sankyo on 16 November 2022. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
Competing interests disclosure
The authors have no competing interests or relevant affiliations with any organization or entity with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
Writing disclosure
No writing assistance was utilized in the production of this manuscript.
Ethical conduct of research
“The authors state that they have not obtained institutional review board approval as this manuscript doesn't require it. Informed consent is not required as there were no human participants involved”.
Open access
This work is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/
Papers of special note have been highlighted as: • of interest; •• of considerable interest
References
- 1. . Pigmented villonodular synovitis: current concepts about diagnosis and management. Future Oncol. 9(10), 1515–1531 (2013).
- 2. . WHO Classification of tumours of soft tissue and bone. IARC, Lyon, 2003, 305–310 (2013).
- 3. . Pigmented villonodular synovitis. J. Am. Acad. Orthop. Surg. 14(6), 376–385 (2006).
- 4. Patient-reported symptoms of tenosynovial giant cell tumors. Clin. Ther. 38(4), 778–793 (2016 Apr).
- 5. Tenosynovial giant cell tumour/pigmented villonodular synovitis: outcome of 294 patients before the era of kinase inhibitors. Eur. J. Cancer 51(2), 210–217 (2015).
- 6. Verschoor ABovee Jvan de Sande MGelderblom H (Eds). Incidence and demographics of giant cell containing tumors in the Netherlands: a nationwide pathology database study. Presented at: Paper presented at Annual Meeting of the Connective Tissue Oncology Society. Salt Lake City, UT, USA, 4–7 November 2015.
- 7. . Pigmented villonodular synovitis and tenosynovitis: a clinical epidemiologic study of 166 cases and literature review. Medicine (Baltimore). 59(3), 223–238 (1980).
- 8. . Long-term follow-up results of primary and recurrent pigmented villonodular synovitis. Rheumatology 53(11), 2063–2070 (2014).
- 9. Pigmented villonodular synovitis: a retrospective multicenter study of 237 cases. PLOS ONE 10(3), e0121451 (2015).
- 10. Higher incidence rates than previously known in tenosynovial giant cell tumors. Acta. Orthop. 88(6), 688–694 (2017). •• A study using the Dutch Pathology Registry (PALGA) to estimate incidence of tenosynovial giant cell tumors, diffuse-type and localized, as well as those affecting digits, concluded that tenosynovial giant cell tumors are not as rare as previously thought.
- 11. . Does CSF1 overexpression or rearrangement influence biological behaviour in tenosynovial giant cell tumours of the knee? Histopathology 74(2), 332–340 (2019).
- 12. . Tenosynovial giant cell tumour, localized type. In: WHO classification of tumours of soft tissue and bone (4th Edition). Fletcher CDMBridge JAHogendoorn PCWMertens F (Eds). WHO Press, Geneva, Switzerland, 5, 100–101 (2013).
- 13. . Giant cell tumor of the tendon sheath (nodular tenosynovitis). A study of 207 cases to compare the large joint group with the common digit group. Cancer 57(4), 875–884 (1986).
- 14. . What affects the recurrence and clinical outcome of pigmented villonodular synovitis? Clin. Orthop. Relat. Res. 450, 172–178 (2006).
- 15. . Localized and diffuse forms of tenosynovial giant cell tumor (formerly giant cell tumor of the tendon sheath and pigmented villonodular synovitis). Orthop. Traumatol. Surg. Res. 103(1S), S91–S97 (2017). •• A deep dive into localized and diffuse types of tenosynovial giant cell tumors outlining clinical presentation and treatment strategies.
- 16. Long term term follow-up of tyrosine kinase inhibitors treatments in inoperable or relapsing diffuse type tenosynovial giant cell tumors (dTGCT). PLOS ONE 15(5), e0233046 (2020).
- 17. Long-term efficacy of imatinib mesylate in patients with advanced tenosynovial giant cell tumor. Sci. Rep. 9(1), 14551 (2019).
- 18. Structure-guided blockade of CSF1R kinase in tenosynovial giant-cell tumor. N. Engl. J. Med. 373(5), 428–437 (2015).
- 19. . Diffuse-type tenosynovial giant cell tumour: current treatment concepts and future perspectives. Eur. J. Cancer 63, 34–40 (2016).
- 20. Pexidartinib versus placebo for advanced tenosynovial giant cell tumour (ENLIVEN), a randomised Phase III trial. Lancet 394(10197), 478–487 (2019).
- 21. CSF1R inhibition with emactuzumab in locally advanced diffuse-type tenosynovial giant cell tumours of the soft tissue: a dose-escalation and dose-expansion Phase I study. Lancet Oncol. 16(8), 949–956 (2015).
- 22. Nilotinib in locally advanced pigmented villonodular synovitis: a multicentre, open-label, single-arm, phase 2 trial. Lancet Oncol. 19(5), 639–648 (2018).
- 23. The diffuse-type tenosynovial giant cell tumor (dt-TGCT) patient journey: a prospective multicenter study. Orphanet J. Rare Dis. 16(1), 191 (2021). •• A prospective observational study using a multinational tenosynovial giant cell tumor registry to collect demographic and clinical characteristics as well as quality of life information.
- 24. Daiichi Sankyo Inc. TURALIO (pexidartinib). [package insert] (2020). Available at: www.accessdata.fda.gov/drugsatfda_docs/label/2020/211810s003lbl.pdf
- 25. . ENLIVEN study: pexidartinib for tenosynovial giant cell tumor (TGCT). Future Oncol. 16(25), 1875–1878 (2020).
- 26. . Update on tenosynovial giant cell tumor, an inflammatory arthritis with neoplastic features. Front. Immunol. 13, 820046 (2022).
- 27. . Soft tissue special issue: giant cell-rich lesions of the head and neck region. Head Neck Pathol. 14(1), 97–108 (2020).
- 28. National Comprehensive Cancer Network. Soft Tissue Sarcoma (Version 1.2022) (2022). www.nccn.org/professionals/physician_gls/PDF/sarcoma.pdf
- 29. System and method for generating de-identified health care data. In: Google Patents. (2004). https://patents.google.com/patent/US7865376B2/en
- 30. System and method for generating de-identified health care data. In: Google Patents. (2008). https://patents.google.com/patent/US7376677B2/en
- 31. . Anonymously linking a plurality of data records. In: Google Patents. (2002). https://patents.google.com/patent/US6397224B1/en
- 32. . Real-world treatment patterns of metastatic non-small-cell lung cancer patients receiving epidermal growth factor receptor tyrosine kinase inhibitors. Cancer Med. 12(1), 159–169 (2023).
- 33. . Burden of illness associated with tenosynovial giant cell tumors. Clin. Ther. 40(4), 593–602.e591 (2018). • A retrospective cohort analysis using administrative claims data from the Optum Research Database which included adult commercial and Medicare Advantage patients to determine treatment burden and healthcare resource utilization in tenosynovial giant cell tumor patients.
- 34. The economic burden of tenosynovial giant cell tumors among employed workforce in the United States. J. Occup. Environ. Med. 63(4), e197–e202 (2021). • A retrospective cohort analysis using the OptumHealth Care Solutions Database to determine healthcare resource utilization and associated workplace productivity loss among tenosynovial giant cell tumor patients.
- 35. Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am. J. Epidemiol. 173(6), 676–682 (2011).
- 36. Statistical Analysis System, SAS. SAS Release 9.4 for windows. SAS Institute Inc., Cary, NC, USA (2013).
- 37. Tenosynovial giant cell tumor: incidence, prevalence, patient characteristics, and recurrence. a registry-based cohort study in Denmark. J. Rheumatol. 44(10), 1476–1483 (2017). •• A linked database study from four nationwide Danish registries with pathology data aimed at identifying patient characteristics, treatment modalities and recurrence for adults with diffuse and localized tenosynovial giant cell tumors.
- 38. U.S. Census Bureau, Population Division. Annual Estimates of the Resident Population for Selected Age Groups by Sex for the United States: April 1, 2010 to July 1, 2019 (NC-EST2019-AGESEX) (2022). www.census.gov/data/tables/time-series/demo/popest/2010s-national-detail.html
- 39. . Surgical treatment patterns, healthcare resource utilization, and economic burden in patients with tenosynovial giant cell tumor who underwent joint surgery in the United States. J. Health Econ. Outcomes Res. 9(1), 68–74 (2022). •• A retrospective analysis using the OptumHealth Care Solutions database to determine surgical patterns and associated economic burden among tenosynovial giant cell tumor patients.
- 40. . Pigmented villonodular synovitis: a comprehensive review and proposed treatment algorithm. JBJS Rev. 4(7), e3 (2016). • A review of treatments for pigmented villonodular synovitis, also known as diffuse type tenosynovial giant cell tumor.
- 41. . Giant cell tumour of tendon sheath of the digits. A systematic review. Hand (N Y). 6(3), 244–249 (2011).
- 42. A multidisciplinary approach to giant cell tumors of tendon sheath and synovium--a critical appraisal of literature and treatment proposal. J. Surg. Oncol. 107(4), 433–445 (2013). • A systematic literature review and expert opinion on treatment strategies for localized and diffuse type tenosynovial giant cell tumors.
- 43. . Outcomes after excision of pigmented villonodular synovitis of the knee. Clin. Orthop. Relat. Res. 467(11), 2852–2858 (2009).
- 44. . Pigmented villonodular synovitis of the hip: 2- to 23-year followup study. Clin. Orthop. Relat. Res. 388(7), 187–199 (2001).
- 45. Surgical outcomes of patients with diffuse-type tenosynovial giant-cell tumours: an international, retrospective, cohort study. Lancet Oncol. 20(6), 877–886 (2019).
- 46. . Treatment patterns of tenosynovial giant cell tumor among commercially insured patients: a retrospective claims analysis. J. Clin. Oncol. 36(suppl. 15), e18737 (2018).
- 47. . The patient perspective on the impact of tenosynovial giant cell tumors on daily living: crowdsourcing study on physical function and quality of life. Interact. J. Med. Res. 7(1), e4 (2018).
- 48. Economic burden and health-related quality of life in tenosynovial giant-cell tumour patients in Europe: an observational disease registry. Orphanet J. Rare Dis. 16(1), 294 (2021).
- 49. . Pexidartinib in the management of advanced tenosynovial giant cell tumor: focus on patient selection and special considerations. Onco Targets Ther. 15, 53–66 (2022).
- 50. . Randomized controlled trials and real-world data: differences and similarities to untangle literature data. Rheumatology 57(7 Suppl. 57), vii54–vii58 (2018).
- 51. Long-term outcomes of pexidartinib in tenosynovial giant cell tumors. Cancer 127(6), 884–893 (2021).
- 52. Novartis. GLEEVEC (imatinib mesylate). [package insert] (2016). Available at: www.accessdata.fda.gov/drugsatfda_docs/label/2016/021588s047lbl.pdf
- 53. Novartis. TASIGNA (nilotinib). [package insert] (2017). Available at: www.accessdata.fda.gov/drugsatfda_docs/label/2017/022068s024lbl.pdf
- 54. Bayer. NEXAVAR (sorafenib). [package insert] (2018). Available at: www.accessdata.fda.gov/drugsatfda_docs/label/2018/021923s020lbl.pdf
- 55. Pfizer. SUTENT (sunitinib malate). [package insert] (2017). Available at: www.accessdata.fda.gov/drugsatfda_docs/label/2017/021938s033lbl.pdf
- 56. Long term term follow-up of tyrosine kinase inhibitors treatments in inoperable or relapsing diffuse type tenosynovial giant cell tumors (dTGCT). PLOS ONE 15(5), e0233046 (2020).
- 57. 2-Year Observational Analysis of Patient-Reported Outcomes and Treatment Strategies from the Tenosynovial Giant Cell Tumor Observational Platform Project (TOPP) Registry. Oncologist (2022) (unpublished).
- 58. Randomized placebo-controlled double-blind phase II study of zaltoprofen for patients with diffuse-type and unresectable localized tenosynovial giant cell tumors: the REALIZE study. Front Oncol. 12, 900010 (2022).
