Digital wearable insole-based identification of knee arthropathies and gait signatures using machine learning.

Gait is impaired in musculoskeletal conditions, such as knee arthropathy. Gait analysis is used in clinical practice to inform diagnosis and monitor disease progression or intervention response. However, clinical gait analysis relies on subjective visual observation of walking as objective gait analysis has not been possible within clinical settings due to the expensive equipment, large-scale facilities, and highly trained staff required. Relatively low-cost wearable digital insoles may offer a solution to these challenges. In this work, we demonstrate how a digital insole measuring osteoarthritis-specific gait signatures yields similar results to the clinical gait-lab standard. To achieve this, we constructed a machine learning model, trained on force plate data collected in participants with knee arthropathy and controls. This model was highly predictive of force plate data from a validation set (area under the receiver operating characteristics curve [auROC] = 0.86; area under the precision-recall curve [auPR] = 0.90) and of a separate, independent digital insole dataset containing control and knee osteoarthritis subjects (auROC = 0.83; auPR = 0.86). After showing that digital insole-derived gait characteristics are comparable to traditional gait measurements, we next showed that a single stride of raw sensor time-series data could be accurately assigned to each subject, highlighting that individuals using digital insoles can be identified by their gait characteristics. This work provides a framework for a promising alternative to traditional clinical gait analysis methods, adds to the growing body of knowledge regarding wearable technology analytical pipelines, and supports clinical development of at-home gait assessments, with the potential to improve the ease, frequency, and depth of patient monitoring.

The way we walk ­ our 'gait' ­ is a key indicator of health. Gait irregularities like limping, shuffling or a slow pace can be signs of muscle or joint problems. Assessing a patient's gait is therefore an important element in diagnosing these conditions, and in evaluating whether treatments are working. Gait is often assessed via a simple visual inspection, with patients being asked to walk back and forth in a doctor's office. While quick and easy, this approach is highly subjective and therefore imprecise. 'Objective gait analysis' is a more accurate alternative, but it relies on tests being conducted in specialised laboratories with large-scale, expensive equipment operated by highly trained staff. Unfortunately, this means that gait laboratories are not accessible for everyday clinical use. In response, Wipperman et al. aimed to develop a low-cost alternative to the complex equipment used in gait laboratories. To do this, they harnessed wearable sensor technologies ­ devices that can directly measure physiological data while embedded in clothing or attached to the user. Wearable sensors have the advantage of being cheap, easy to use, and able to provide clinically useful information without specially trained staff. Wipperman et al. analysed data from classic gait laboratory devices, as well as 'digital insoles' equipped with sensors that captured foot movements and pressure as participants walked. The analysis first 'trained' on data from gait laboratories (called force plates) and then applied the method to gait measurements obtained from digital insoles worn by either healthy participants or patients with knee problems. Analysis of the pressure data from the insoles confirmed that they could accurately predict which measurements were from healthy individuals, and which were from patients. The gait characteristics detected by the insoles were also comparable to lab-based measurements ­ in other words, the insoles provided similar type and quality of data as a gait laboratory. Further analysis revealed that information from just a single step could reveal additional information about the subject's walking. These results support the use of wearable devices as a simple and relatively inexpensive way to measure gait in everyday clinical practice, without the need for specialised laboratories and visits to the doctor's office. Although the digital insoles will require further analytical and clinical study before they can be widely used, Wipperman et al. hope they will eventually make monitoring muscle and joint conditions easier and more affordable.

Monoclonal antibodies against GFRα3 are efficacious against evoked hyperalgesic and allodynic responses in mouse join pain models but, one of these, REGN5069, was not effective against pain in a randomized, placebo-controlled clinical trial in patients with osteoarthritis pain.

The artemin-GFRα3 signaling pathway has been implicated in various painful conditions including migraine, cold allodynia, hyperalgesia, inflammatory bone pain, and mouse knees contain GFRα3-immunoreactive nerve endings. We developed high affinity mouse (REGN1967) and human (REGN5069) GFRα3-blocking monoclonal antibodies and, following in vivo evaluations in mouse models of chronic joint pain (osteoarthritic-like and inflammatory), conducted a first-in-human phase 1 pharmacokinetics (PK) and safety trial of REGN5069 (NCT03645746) in healthy volunteers, and a phase 2 randomized placebo-controlled efficacy and safety trial of REGN5069 (NCT03956550) in patients with knee osteoarthritis (OA) pain. In three commonly used mouse models of chronic joint pain (destabilization of the medial meniscus, intra-articular monoiodoacetate, or Complete Freund's Adjuvant), REGN1967 and REGN5069 attenuated evoked behaviors including tactile allodynia and thermal hyperalgesia without discernably impacting joint pathology or inflammation, prompting us to further evaluate REGN5069 in humans. In the phase 1 study in healthy subjects, the safety profiles of single doses of REGN5069 up to 3000 mg (intravenous) or 600 mg (subcutaneous) were comparable to placebo; PK were consistent with a monoclonal antibody exhibiting target-mediated disposition. In the phase 2 study in patients with OA knee pain, two doses of REGN5069 (100 mg or 1000 mg intravenous every 4 weeks) for 8 weeks failed to achieve the 12-week primary and secondary efficacy endpoints relative to placebo. In addition to possible differences in GFRα3 biology between mice and humans, we highlight here differences in experimental parameters that could have contributed to a different profile of efficacy in mouse models versus human OA pain. Additional research is required to more fully evaluate any potential role of GFRα3 in human pain.

Effect of High-Dose Intravitreal Aflibercept, 8 mg, in Patients With Neovascular Age-Related Macular Degeneration: The Phase 2 CANDELA Randomized Clinical Trial.

Importance: Aflibercept, 8 mg, may have greater therapeutic benefits compared with aflibercept, 2 mg, in patients with neovascular age-related macular degeneration (nAMD), including potentially improved outcomes and decreased treatment burden. Objective: To assess safety and efficacy of aflibercept, 8 mg, in patients with nAMD. Design, Setting, and Participants: The CANDELA trial was a phase 2, randomized, single-masked, open-label, 44-week clinical trial conducted in the US. Treatment-naive patients with active subfoveal choroidal neovascularization secondary to nAMD and a best-corrected visual acuity score of 78 to 24 letters (approximately 20/32 to 20/320) in the study eye were enrolled between November 2019 and November 2021. Interventions: Eligible participants were randomized 1:1 to receive 3 monthly doses of 8 mg (70 µL) or 2 mg (50 µL) of aflibercept followed by doses at weeks 20 and 32. Main Outcomes and Measures: Coprimary end points were the proportion of eyes without fluid (absence of intraretinal and subretinal fluid) in the central subfield at week 16 and safety. Results: All 106 eligible eyes were randomized to receive aflibercept, 8 mg (n = 53), or aflibercept, 2 mg (n = 53). Overall, 66 participants (62.3%) were female. The proportion of eyes without fluid in the central subfield with 8-mg vs 2-mg aflibercept was 50.9% (n = 27) vs 34.0% (n = 18) (difference, 17.0 [95% CI, -1.6 to 35.5] percentage points; P = .08) at week 16 and 39.6% (n = 21) vs 28.3% (n = 15) (difference, 11.3 [95% CI, -6.6 to 29.2] percentage points; nominal P = .22) at week 44. At week 44, mean (SE) change in central retinal thickness was -159.4 (16.4) vs -137.2 (22.8) µm with 8 mg vs 2 mg of aflibercept, respectively (least squares mean difference, -9.5 [95% CI, -51.4 to 32.4]; nominal P = .65) and mean (SE) change in best-corrected visual acuity score was +7.9 (1.5) vs +5.1 (1.5) letters (least squares mean difference, +2.8 [95% CI, -1.4 to +7.0]; nominal P = .20). No differences in safety profiles between the groups were observed. Conclusions and Relevance: Although aflibercept, 8 mg, did not achieve the primary efficacy end point at week 16 at the 2-sided significance level of 5%, the observed trends in anatomic and visual improvements over 44 weeks with aflibercept, 8 mg, indicate potential additional therapeutic benefit over aflibercept, 2 mg. No new safety signals were observed over 44 weeks. These findings support further evaluation of aflibercept, 8 mg, in pivotal trials of exudative retinal diseases including nAMD and diabetic macular edema. Trial Registration: ClinicalTrials.gov Identifier: NCT04126317.

A pharmacodynamically guided dose selection of PF-00337210 in a phase I study in patients with advanced solid tumors.

PURPOSE: PF-00337210 is an oral, highly selective vascular endothelial growth factor receptor (VEGFR) inhibitor. We evaluated a composite of biomarkers in real time to identify the recommended phase 2 dose (RP2D) and preliminary anticancer activity of PF-00337210. PATIENTS AND METHODS: Patients (Pts) with advanced cancers were treated once (QD) or twice daily (BID) with escalating doses. Acute effects on tumor perfusion and vascularity were assessed using DCE-MRI, weekly BP readings, soluble VEGFR-2, and hemoglobin levels. RESULTS: Forty-six pts were treated with 0.67-9 mg QD and 4-6 mg BID of PF-00337210. Nineteen pts (41%) previously received VEGF/VEGFR inhibitors. Two pts had dose-limiting toxicity (DLT) at 9 mg QD (troponin I increase and hypertension). The MTD at QD dose was 8 mg. Common drug-related adverse events were hypertension, fatigue, proteinuria, and nausea. Hypertension incidence and intensity corresponded with dose, but was well controlled with medication. Two confirmed partial responses and minor regressions (>10 to <30% reduction in target lesions) were noted. Complete DCE-MRI was acquired in 21 pts (20 evaluable for vascular response). Ten pts were vascular responders, including 5/6 pts at BID doses. Greatest modulation of soluble VEGFR-2 was at 6 mg BID. The maximum change from baseline in diastolic BP was higher at BID doses. There were no significant differences for systolic BP and hemoglobin levels. CONCLUSIONS: PF-00337210 has profound VEGFR inhibition effects at well-tolerated doses. Antitumor activity and VEGF inhibition effects were observed across BID doses. The RP2D was 6 mg BID.

Cardiovascular events among 1090 cancer patients treated with sunitinib, interferon, or placebo: a comprehensive adjudicated database analysis demonstrating clinically meaningful reversibility of cardiac events.

PURPOSE: To define cardiovascular (CV) risk and reversibility of cardiac events in patients who received sunitinib versus comparator treatment (interferon-alfa or placebo). PATIENTS AND METHODS: We performed a retrospective adjudication of comprehensive CV adverse events (AEs) from two phase 3 trials. Components of the comprehensive CV AE end-point comprised hypertension, symptomatic and asymptomatic left ventricular ejection fraction decreases (SD-LVEF; AD-LVEF) and extent of reversibility, heart-failure symptoms, thromboembolic events, dysrhythmia and CV death. Three cardiologists and one oncologist, blinded to treatment allocation, adjudicated suspected CV AEs in the pooled trial database (N=1090). RESULTS: Incidence rates (IR) for sunitinib versus Interferon-alfa (IFN-α)/placebo were hypertension: 6.9 versus 2.6 (hazard ratio (HR), 3.1; 95% confidence interval (CI), 2.4-4.0); SD-LVEF: 0.4 versus 0.2 (HR, 2.5; 95% CI, 1.0-6.2); AD-LVEF: 1.1 versus 0.8 (HR, 2.1; 95% CI, 1.3-3.4); and composite CV AE end-point: 10.1 versus 4.8 (HR, 2.5; 95% CI, 2.0-3.1), however reversibility, not previously quantified, was found to be clinically meaningful. CONCLUSIONS: Hypertension and SD-LVEF/AD-LVEF were significantly higher with sunitinib versus IFN-α/placebo. Among patients who experienced a cardiac event, symptomatic and asymptomatic instances of decreased cardiac dysfunction were adjudicated as reversible in 47 of 83 (56%) and 17 of 30 (57%), respectively. Among sunitinib-treated patients, many were able to resume sunitinib dosing following resolution of events, a finding that is important for clinical care. In comparator groups, symptomatic and asymptomatic instances were adjudicated as reversible in 4 of 6 (66.7%) and 11 of 21 (52%), respectively. Thromboembolic, dysrhythmic and/or CV deaths were not significantly higher in sunitinib-treated patients.

Phase II study of irinotecan in combination with temozolomide (TEMIRI) in children with recurrent or refractory medulloblastoma: a joint ITCC and SIOPE brain tumor study.

BACKGROUND: This multicenter phase II study investigated temozolomide + irinotecan (TEMIRI) treatment in children with relapsed or refractory medulloblastoma. METHODS: Patients received temozolomide 100-125 mg/m(2)/day (days 1-5) and irinotecan 10 mg/m(2)/day (days 1-5 and 8-12) every 3 weeks. The primary endpoint was tumor response within the first 4 cycles confirmed ≥4 weeks and assessed by an external response review committee (ERRC). In a 2-stage Optimum Simon design, ≥6 responses in the first 15 evaluable patients were required within the first 4 cycles for continued enrollment; a total of 19 responses from the first 46 evaluable patients was considered successful. RESULTS: Sixty-six patients were treated. Seven responses were recorded during stage 1 and 15 in the first 46 ERRC evaluated patients (2 complete responses and 13 partial responses). The objective response rate during the first 4 cycles was 32.6% (95% confidence interval [CI], 19.5%-48.0%). Median duration of response was 27.0 weeks (7.7-44.1 wk). In 63 patients evaluated by local investigators, the objective response rate was 33.3% (95% CI, 22.0%-46.3%), and 68.3% (95% CI, 55.3%-79.4%) experienced clinical benefit. Median survival was 16.7 months (95% CI, 13.3-19.8). The most common grade 3 treatment-related nonhematologic adverse event was diarrhea (7.6%). Grade 3/4 treatment-related hematologic adverse events included neutropenia (16.7%), thrombocytopenia (12.1%), anemia (9.1%), and lymphopenia (9%). CONCLUSIONS: The planned study primary endpoint was not met. However, its tolerability makes TEMIRI a suitable candidate chemotherapy backbone for molecularly targeted agents in future trials in this setting.

A phase II single-arm study of irinotecan in combination with temozolomide (TEMIRI) in children with newly diagnosed high grade glioma: a joint ITCC and SIOPE-brain tumour study.

A multicenter, two stage phase II study, investigated irinotecan plus temozolomide in children with newly diagnosed high grade glioma. The primary endpoint was tumor response during a two-cycle treatment window, confirmed by external review committee. Patients received oral temozolomide 100 mg/(m(2) day) (days 1-5) and intravenous irinotecan 10 mg/(m(2) day) (days 1-5 and 8-12) for two 21-day cycles (three cycles for patients exhibiting objective tumor response). Standard treatment was then administered according to local investigator choice. In total 17 patients were enrolled and treated by local investigators. However, central pathology review found three patients did not have a diagnosis of high grade glioma and another four patients did not have evaluable disease according to independent central radiological review. The primary endpoint was based on the first ten evaluable patients as determined by the external review committee. Recruitment was stopped for futility after there were no complete or partial responses during the two-cycle treatment window in the first ten evaluable patients. Five patients had stable disease, and five progressed. Data for secondary endpoints including; time to tumor progression, time to treatment failure, and overall survival is reported. The safety profile of the treatment showed the combination was tolerable with two patients (11.8 %) having grade three nausea, and one (5.9 %) experiencing a grade four neutropenia, leading to permanent discontinuation from adjuvant treatment. Irinotecan plus temozolomide, although well tolerated did not improve outcome over historical controls in this setting.

Treatment of osteoarthritis with continuous versus intermittent celecoxib.

OBJECTIVE: To determine whether "continuous" celecoxib is more efficacious than "intermittent" use in preventing osteoarthritis (OA) flares of the knee and/or hip. METHODS: A double-blind, randomized, multicenter international study comparing efficacy and safety of continuous (daily) versus intermittent (as required during predefined OA flare) celecoxib 200 mg/day in 858 subjects, aged 18-80 years. The study consisted of 3 periods: (I) screening/washout visit; (II) open-label run-in with celecoxib; and (III) 22-week blinded treatment. Only subjects whose OA flares resolved in Period 2 (without subsequent flare) were randomized. The primary endpoint, number of flares per time of exposure during Period III (number of flares per month), was compared using analysis of variance with treatment as the independent variable. Acetaminophen was available as rescue medication. RESULTS: Of 875 subjects randomized to treatment, 858 subjects received treatment. At randomization > 70% were female; mean age 58.6 years; mean disease duration 6.5 years; total Western Ontario and McMaster Universities Osteoarthritis Index mean score 25.8; ~45% had hypertension; and ~20% were using aspirin (for cardiovascular prophylaxis). Subjects receiving continuous treatment reported 42% fewer OA flares/month than intermittent users (p < 0.0001) or 2.0 fewer OA flares over 22 weeks. Statistical and clinically meaningful benefits in secondary outcomes were also evident with continuous treatment. There were no differences in adverse events (AE) or new-onset/aggravated hypertension. CONCLUSION: Continuous treatment with celecoxib 200 mg/day was significantly more efficacious than intermittent use in preventing OA flares of the hip and knee, without an increase in overall AE, including gastrointestinal disorders and hypertension, during 22 weeks of treatment. ClinicalTrials.gov identifier NCT00139776.

Five-year analysis of the prevention of colorectal sporadic adenomatous polyps trial.

OBJECTIVES: Subjects in the Prevention of Colorectal Sporadic Adenomatous Polyps (PreSAP) trial (PRESAP/NCT00141193/www.clinicaltrials.gov) were studied to determine efficacy and safety at a year 5 assessment. METHODS: In this randomized, placebo-controlled, double-blind trial, 1,561 subjects with diagnosed colorectal adenomas removed within 3 months of the study's initiation were assessed after ~ 3 years on celecoxib followed by 2 years off. Studied in 107 primary and secondary care settings, subjects were stratified by cardioprotective aspirin use and randomized to receive orally 400 ng celecoxib (933 subjects) or placebo (628 subjects) once daily. Efficacy was measured by colonoscopy at years 1, 3, and 5, and safety was measured by investigators for the on-treatment period and collected by subject self-report over 2 years post-treatment. RESULTS: At year 5, the primary outcome measure was the rate of new adenomas measured cumulatively from baseline. This rate was statistically significantly lower in the celecoxib group (51.4%) than in the placebo group (57.5%; P<0.001). Similarly, the cumulative rate of new advanced adenomas was significantly lower in the celecoxib group (10.0%) than in the placebo group (13.8%; P=0.007). However, the year 5 interval measure, which was not cumulative and did not take the rates of previous years into account, showed that after 2 years off treatment, the celecoxib group (27.0%) was 1.66 times more likely to have new adenomas than the placebo group (16.3%; P<0.0001). Similarly, the percentage of patients with new advanced adenomas was significantly higher in the celecoxib group (5.0%) than in the placebo group (3.8%) (P=0.0072). The evaluation of safety from baseline through year 5 indicated that the risks of serious cardiac disorders (relative risk (RR) 1.66; 95% confidence interval (CI) 1.01-2.73), selected renal/hypertension events (RR 1.35; 95% CI 1.09-1.68), and general vascular (RR 1.34; 95% CI 1.08-1.68) and cardiac disorders (RR 1.59; 95% CI 1.12-2.26) were higher in those taking celecoxib than in those on placebo. CONCLUSIONS: The year 5 cumulative measures of the incidence of new and advanced adenomas were significantly lower in the celecoxib group than in the placebo group, but the year 5 interval rates of these measures were significantly lower in the placebo group than the celecoxib group, perhaps suggesting a release of cyclooxygenase-2 inhibition. Consistent with what has been previously reported, increased risk of renal/hypertension events and cardiac disorders associated with celecoxib therapy mandates caution in patient selection.

A novel composite endpoint to evaluate the gastrointestinal (GI) effects of nonsteroidal antiinflammatory drugs through the entire GI tract.

OBJECTIVE: Nonsteroidal antiinflammatory drugs (NSAID) not only cause damage to the upper gastrointestinal (GI) tract but also affect the lower GI tract. To date, there is no endpoint that evaluates serious GI events in the entire GI tract. The objective of this report is to introduce a novel composite endpoint that measures damage to the entire GI tract - clinically significant upper and lower GI events (CSULGIE) - in patients with NSAID-induced GI damage. METHODS: We reviewed the data from largescale, multicenter, randomized, clinical trials on lower GI toxicity associated with NSAID use. The rationale for using CSULGIE as a primary endpoint in 2 ongoing trials - the Celecoxib vs Omeprazole and Diclofenac for At-risk Osteoarthritis (OA) and Rheumatoid Arthritis (RA) Patients (CONDOR) trial and the Gastrointestinal Randomized Events and Safety Open-Label NSAID Study (GI-REASONS) - is also discussed. RESULTS: Previous randomized trials focused primarily on damage to the upper GI tract and often neglected the lower GI tract. The CSULGIE endpoint extends the traditional "perforation, obstruction, and bleeding" assessment of upper GI complications by including events in the lower GI tract (small/large bowel) such as perforation, bleeding, and clinically significant anemia. CONCLUSION: By providing clinicians with a new, descriptive language for adverse events through the entire GI tract, the CSULGIE endpoint has the potential to become a standard tool for evaluating the GI effects of a range of therapies.

Five-year efficacy and safety analysis of the Adenoma Prevention with Celecoxib Trial.

The Adenoma Prevention with Celecoxib Trial examined the efficacy and safety of the cyclooxygenase (Cox)-2 inhibitor, celecoxib, for sporadic colorectal adenoma prevention in patients at high risk for colorectal cancer. The trial randomized 2,035 subjects to receive either placebo, celecoxib 200 mg twice daily, or celecoxib 400 mg twice daily. The primary study safety and efficacy analyses involved 3 years of treatment. The results showed significant antitumor effect but also indicated increased cardiovascular adverse events in patients treated with celecoxib compared with placebo. A total of 933 patients participated in an extension of the Adenoma Prevention with Celecoxib Trial, with a planned total treatment and surveillance duration of 5 years. Study medication was stopped early, resulting in a median treatment duration of 3.1 years for those with a year 5 colonoscopy. Patients treated on the placebo arm had a cumulative adenoma incidence of 68.4% over 5 years of observation. This figure was 59.0% (P < 0.0001) for those receiving low-dose celecoxib, and 60.1% (P < 0.0001) for those receiving high-dose celecoxib. The cumulative incidence of advanced adenomas over 5 years was 21.3% of those taking placebo, 12.5% (P < 0.0001) of those taking low dose celecoxib and 15.8% (P < 0.0001) of those taking high-dose celecoxib. Investigator reported treatment emergent adverse events were similar across all treatment groups for categories including renal and hypertensive events and gastrointestinal ulceration and hemorrhage events. For a category composed of cardiovascular and thrombotic events, the risk relative to placebo was 1.6 (95% confidence interval, 1.0, 2.5) for those using 200 mg twice daily celecoxib and 1.9 (95% confidence interval, 1.2, 3.1) for those using 400 mg twice daily celecoxib. Secondary analysis showed an interaction between a baseline history of atherosclerotic heart disease and study drug use with respect to cardiovascular and thrombotic adverse events (P = 0.004). These results confirm the inhibitory effect of celecoxib on colorectal adenoma formation, and provide additional safety data indicating an elevated risk for cardiovascular and thrombotic adverse events, particularly for patients with preexisting atherosclerotic heart disease.

Cytochrome P450 2C9 variants influence response to celecoxib for prevention of colorectal adenoma.

BACKGROUND & AIMS: Variants in the cytochrome P450 2C9 (CYP2C9) gene are associated with impaired metabolism of celecoxib. We examined the influence of CYP2C9*2 (R144C) and CYP2C9*3 (I359L) variants on dose-related response or toxicity in a randomized trial of celecoxib. METHODS: We identified individuals with CYP2C9*2 and CYP2C9*3 genotypes (>or=1 variant allele) in the Adenoma Prevention with Celecoxib trial. Following adenoma removal, patients were assigned randomly to groups given placebo or low-dose (200 mg twice daily) or high-dose (400 mg twice daily) celecoxib and underwent follow-up colonoscopies at 1 and/or 3 years. RESULTS: Among 1660 patients, 21% were CYP2C9*2, and 12% were CYP2C9*3 genotypes. Overall, celecoxib was associated with a dose-dependent reduction in adenoma, compared with placebo, with relative risks (RR) of 0.65 (95% confidence interval [CI]: 0.56-0.76) for the low-dose and 0.54 (95% CI: 0.46-0.63) for the high-dose groups. However, the additional protective effect of the high dose, compared with the low-dose, was observed only in those with CYP2C9*3 genotypes (RR, 0.51; 95% CI: 0.30-0.87). The high dose, compared with low dose, was not associated with significant risk reduction among those with CYP2C9*2 (RR, 0.83; 95% CI: 0.57-1.21) or wild-type (RR, 0.89; 95% CI: 0.72-1.11) genotypes. Compared with placebo, a higher incidence of cardiovascular events was associated with both doses among patients with wild-type genotypes but only with the high dose among patients with variant genotypes. CONCLUSIONS: The greater efficacy of high-dose celecoxib, compared with the low-dose, in preventing colorectal adenoma appears confined to individuals with slow metabolizer (CYP2C9*3) genotypes. Genetic variability influences susceptibility to the potential benefits and hazards of celecoxib.