Table 2.

Summary of Studies in Patients With TGCT in Various Sites

Study (No. of pts)	Study Design (Tx Period), LOE	Dx, F/U	Tx	Efficacya	Complicationsa
Surgery
Capellen et al25 (n = 105)	Retrospective cohort (1996-2014) LOE: III	PVNS in knee (n = 58), feet (n = 16), hand (n = 11), ankle (n = 9), hip (n = 4), elbow (n = 2), lower calf (n = 2), sacral joints (n = 1), upper calf (n = 1), shoulder (n = 1) Diffuse PVNS seen in 66 (54%) lesions F/U: 71b mo (n = 103; 2 pts lost to F/U)	All pts had open Sv (n = 120 resections); some pts also had arthroscopy for diagnostic reasons	Recurrence all pts: 22 of 120 (18%) resections Recurrence at knee: 8 of 36 (22%) and 3 of 28 (11%) nodular lesion Persistent tumors at last f/u: 6 of 103 pts (5.8%) Recurrence in pts who underwent RSO: 7 of 27 pts (32%)	Postop: wound revisions due to hematomas (n = 2); necrosis of femoral condyle (n = 1); peroneal nerve palsy (n = 1); deep infection (n = 1); instability of collateral ligament at knee (n = 1)
Ma et al26 (n = 75)	Retrospective cohort (2000-2010) LOE: III	PVNS in knee (n = 52), hip (n = 18), ankle (n = 4), hand (n = 1) Localized PVNS (n = 8) or diffuse PVNS (n = 67) F/U: 40 moc (n = 60 pts kept all appointments)	All pts underwent open or arthroscopic Sv alone (n = 41 pts; n = 43 joints) or with arthroplasty (n = 34; n = 38 joints)	Recurrence on MRI: 17 joints with diffuse PVNS	No postop complications or infection
Mastboom et al27 (n = 941; n = 930 received tx at participating tertiary center; n = 823 with complete survival data)	Retrospective cohort (1990-2017) LOE: III	Localized TGCT in knee (n = 633), ankle (n = 119), foot (n = 58), hip (n = 37), hand (n = 33), wrist (n = 24), elbow (n = 14), shoulder (n = 9), other (n = 14) F/U: 37b mo	Type of tx known for 930 pts: Open resection (n = 675), arthroscopic resection (n = 140), wait and see (n = 64), resection NS (n = 30), endoprosthetic reconstruction (n = 21)	Recurrence (n = 823): 12% LRFS for all pts: 3-yr, 88% (n = 388); 5-yr, 83% (n = 231); 10 yr, 79% (n = 66) RFS for tx naïve pts: 3-yr, 90% (n = 372); 5-yr, 86% (n = 223); 10-yr, 82% (n = 63) RFS: open (87%) versus arthroscopic surgery (80%) (P = 0.04); statistical significance lost in multivariate analysis Symptoms prior to treatment (n = 663-767) versus final F/U (n = 522-525): pain (73% versus 25%); swelling (66% versus 12%); joint stiffness (10% versus 4%); limited ROM (16% versus 5%)	Postop complications after surgery at tertiary center (n = 763): superficial wound infection (n = 11; 1%), deep wound infection (n = 1; 0.1%), joint stiffness (n = 5; 0.7%), hemorrhage (n = 1; 0.1%), neurovascular damage (n = 3; 0.4%), thrombosis (n = 3; 0.4%), other (n = 10; 1%)
Mastboom et al11 (n = 1192; n = 966 received surgery as primary tx and had complete data)	Retrospective cohort (1990-2017) LOE: III	Diffuse TGCT in knee (n = 758), hip (n = 124), ankle (n = 162), foot (n = 63), shoulder (n = 15), elbow (n = 17), wrist (n = 25), hand (n = 13), other (n = 15) F/U: 54b mo	1-staged open Sv (n = 628), 2-staged open Sv (n = 187), arthroscopic Sv (n = 159), wait and see (n = 76), Sv not specified (n = 47)	First local recurrence (n = 966): 44% Total no. of recurrences (n = 425): 1 (63%); 2 (20%); ≥ 3 (17%) No evidence of disease at last F/U (n = 587): 66% LRFS for pts who received primary tx (n = 966): 3-yr, 62% (n = 474); 5-yr, 55% (n = 297); 10-yr, 40% (n = 89) LRFS for tx-naïve pts (n = 758): 3-yr, 70% (n = 372); 5-yr, 64% (n = 227); 10-yr, 50% (n = 70) Univariate analysis of 5-yr LRFS: Open (66%) versus arthroscopic surgery (54%) (P = 0.03); statistical significance lost in multivariate analysis Symptoms prior to treatment (n = 161–738) versus final F/U (n = 92–233): Pain (76% versus 37%); swelling (75% versus 24%); joint stiffness (21% versus 17%); limited ROM (28% versus 19%)	Postop complications after surgery at tertiary center (n = 906): superficial wound infection (n = 15; 2%), deep wound infection (n = 10; 1%), joint stiffness (n = 32; 4%), hemorrhage (n = 7; 1%), neurovascular damage (n = 15; 2%), thrombosis (n = 1; 0.1%), other (n = 25; 3%)
Palmerini et al28 (n = 294)d	Retrospective cohort (1998-2008) LOE: III	TGCT in knee (60%), ankle (16%), hip (11%), other sites (13%) Localized TGCT (n = 90) or diffuse TGCT (n = 196) F/U: 4.4b yr	Open Sv (n = 171) or arthroscopic Sv (n = 66) None of the pts had EBRT or other local adjuvant tx Medical tx: Neoadjuvant imatinib (n = 2); disphosphonate (n = 1)	Local failure: All pts (28%); pts w/diffuse TGCT (36%); pts w/localized TGCT (14%) 5-yr LFFS: All pts (66%) 5-yr LFFS after arthroscopy versus open surgery: Localized TGCT (84% versus 72%; P = 0.4); diffuse TGCT (59% versus 61%; P = 0.8)	NR
Xie et al29 (n = 237)	Retrospective cohort (2005-2014) LOE: III	PVNS in knee (n = 175), hip (n = 43), ankle (n = 8), wrist (n = 6), shoulder (n = 2), elbow (n = 2), finger (n = 1) Localized or diffuse PVNS status: NR F/U: NR	Arthroscopic Sv (n = 129); open Sv (n = 108)	Recurrence: all pts (20%); knee (24%); hip (7%)	NR
Drug therapies
Cassier et al30 (n = 29 enrolled; n = 28 evaluable for efficacy; n = 25 evaluable for safety; n = 17 in dose expansion cohort)	Phase 1 trial (2012-2014) LOE: II Primary objective: evaluate safety and tolerability, determine MTD or OBD	Locally advanced diffuse TGCT in knee (n = 15), foot or ankle (n = 8), hip (n = 4), wrist (n = 2) F/U: 12b mo	Emactuzumab (1.5 h infusion every 2 wk) 3 + 3 dose-escalation phase followed by expansion phase	MTD not reached ORR: 86% CRs: 7% PRs: 79%	DLTs: None Most common AEs: facial edema (64%), asthenia (56%), pruritus (56%) SAEs (n = 5 pts): periorbital edema, lupus erythematosus, erythema, dermohypodermitis
Gelderblom et al31 (n = 56; n = 51 evaluable for efficacy)	Phase 2 trial (2010-2012) LOE: II Primary end point: proportion of pts progression-free at 12 wk	Inoperable progressive or relapsing PVNS or PVNS only resectable with mutilating surgery in knee (n = 29), ankle or foot (n = 13), hip or femoral neck (n = 7), hand or finger (n = 3), wrist (n = 2), ulna (n = 1), other (n = 1) F/U: 48 mo (n = 50)	Nilotinib (twice per day) 5 pts not evaluable for primary end point because discontinued study tx before week 12 Median duration of tx: 11 mo (IQR, 7.0-12.0)	% pts progression-free at 24 weeks: 90% Estimated mean proportion of pts progression-free at 24 weeks: 88% No CRs or PRs SD: 90% 31 pts (55%) completed 1 year of tx 1-yr PFS: 77%	% with AEs: ≥1 AE (98%); ≥1 TRAE (96%); AEs leading to treatment modification (41%); ≥1 grade 3 TRAE (11%) Grade 3 TRAE: headache, dizziness, hepatic disorders (n = 1); pruritus and toxidermia (n = 1); diarrhea (n = 1); increased γ-glutamyl transferase concentrations (n = 1); anorexia (n = 1); increased headache (n = 1)
Tap et al32 (n = 23 extension phase)	Phase 1 trial (2009-2014) LOE: II Primary objective: clinical benefit	TGCT in knee (n = 15), hip (n = 2), foot (n = 2), ankle (n = 2), elbow (n = 1), forearm (n = 1), metastatic (n = 1), with demonstrated progression within past 1 yr that was recurrent, inoperable, or resectable but requiring extensive surgery F/U: NR	Pexidartinib (once a day) Mean duration of tx: 254 d (range 15-585)	ORR: 52% (12 pts with PRs, 0 CRs) SD: 30% (7 pts) Responses usually occurred within 4 mo of tx start	Any TRAEs: 100% Most common: hair color changes (74%); fatigue (65%); nausea (39%); dysgeusia (26%); periorbital edema (26%); decreased appetite (26%); diarrhea (30%); vomiting (30%)
Tap et al33 (n = 120)  Note: planned sample size of n = 126 for part 1e	Phase 3, double-blind, placebo-controlled, RCT (2015-2018) LOE: I Primary end point: ORR at week 25, based on blinded central MRI	Symptomatic advanced TGCT in knee (n = 73), ankle (n = 21), hip (n = 13), wrist (n = 4), foot (n = 3), shoulder (n = 2), spine (n = 2), elbow (n = 1), finger (n = 1), for whom surgery was not recommended F/U: 22 mob	Part 1 (double-blind phase), twice daily pexidartinib (n = 61) or placebo (n = 59) for 24 wk Part 2, open-label pexidartinib at dose of pexidartinib or placebo ended on part 1 (n = 30)	ORR at week 25: Pexidartinib 39% (9 CRs, 15 PRs) versus placebo 0% (P < 0.0001) Difference (pexidartinib—placebo) in ROM change from baseline to week 25: mean + 8.9 ± 3.0 (P = 0.0043)	Pexidartinib versus placebo Most common any grade AEs: hsair color changes (67% versus 3%); fatigue (54% versus 36%); AST increase (39% versus 0%); nausea (38% versus 41%); ALT increase (28% versus 2%); diarrhea (20% versus 25%)
Verspoor et al34 (n = 62; n = 58 evaluable for efficacy)	Retrospective cohort (NR) LOE: III	Locally advanced, recurrent, or metastatic diffuse TGCT in knee (n = 35), ankle (n = 11), hip (n = 6), foot (n = 4), head and neck (n = 2), wrist (n = 2), shoulder (n = 1), elbow (n = 1) F/U: 52 moc	Imatinib mesylate (once a day) Median duration of tx: 9 mo (IQR 5-26)	ORR: 31% CR: 4% PR: 27% SD: 65% Symptom improvement: 78% Median PFS: 18 mo (IQR 8-55) Median time to best response: 6 mo (range 1-23)	Most common AEs any grade: fatigue (50%); edema/fluid retention (48%); nausea (34%); skin rash/dermatitis (12%); other (26%) Grade 3-4 AEs: edema/fluid retention (2%); fatigue (2%); skin rash/dermatitis (3%); other (5%)

AE = adverse event, ALT = alanine aminotransferase, AST = aspartate aminotransferase, CI = confidence interval, CR = complete response, DLT = dose-limiting toxicity, DTGCT = diffuse-type giant cell tumor, Dx = diagnosis, EBRT = external beam radiation therapy, F/U = follow-up, IQR = interquartile range, KSS = Knee Society Score, LFFS = local failure-free survival, LOE = level of evidence, LRFS = local relapse-free survival; MTD = maximum tolerated dose, NR = not reported, NS = not specified, OA = osteoarthritis, OBD = optimal biologic dose, ORR = objective response rate, PFS = progression-free survival, postop = postoperatively, PR = partial response, pts = patients, PVNS = pigmented villonodular synovitis, RCT = randomized controlled trial, RFS = recurrence-free survival, ROM = range of motion, RSO = radiosynoviorthesis, SAE = serious adverse event, SD = stable disease, Sv = synovectomy, TGCT = tenosynovial giant cell tumor, TRAE = treatment-related adverse event, Tx = treatment

^aPercentage of patients, unless otherwise indicated.

^bMedian.

^cMean or average.

^dStudy excluded patients with clinical history of > 1 local recurrence.

^eStudy protocol was amended to halt enrollment at 120 patients based on Data Monitoring Committee recommendation following cases of mixed or cholestatic hepatoxicity (2 patients in Tap et al, 2019 study, and other cases in pexidartinib's non-TGCT development plan).