Search

Canakinumab is a human IgGκ monoclonal antibody, with high affinity and specificity for IL-1β. The Canakinumab Anti-Inflammatory Thrombosis Outcome Study (CANTOS) trial, evaluating canakinumab for cardiovascular disease, provided the first signal of the potential of IL-1β inhibition on lung cancer incidence reduction. Here, we describe the rationale and design for CANOPY-N, a randomized Phase II trial evaluating IL-1β inhibition with or without immune checkpoint inhibition as neoadjuvant treatment in patients with non-small-cell lung cancer. Patients with stage IB to IIIA non-small-cell lung cancer eligible for complete resection will receive canakinumab or pembrolizumab as monotherapy, or in combination. The primary end point is major pathological response by central review; secondary end points include overall response rate, major pathological response (local review), surgical feasibility rate and pharmacokinetics.Lay abstract A previous study showed that canakinumab reduced the risk of lung cancer in patients with heart disease. Canakinumab blocks an inflammatory protein called IL-1β that is involved in cancer. Anti-cancer drugs used before surgery (‘neo-adjuvant’) can improve the success rate of surgery and may help prevent the cancer from returning. Neo-adjuvant trials help us understand how the drugs work and how they affect cancer. CANOPY-N (NCT03968419) is an ongoing randomized, exploratory, Phase II clinical trial testing canakinumab and pembrolizumab (a different cancer immunotherapy), alone or combined, for patients with early non-small-cell lung cancer. The study will test whether treatment can kill most cancer cells in the surgery sample (‘major pathological response’). It will also investigate other effects on cancer biology, levels of molecules that measure possible clinical benefit (‘biomarkers’) and side effects.

Summary Background Polycythaemia vera is a myeloproliferative neoplasm characterised by excessive proliferation of erythroid, myeloid, and megakaryocytic components in the bone marrow due to mutations in the Janus kinase 2 (JAK2) gene. Ruxolitinib, a JAK 1 and JAK 2 inhibitor, showed superiority over best available therapy in a phase 2 study in patients with polycythaemia vera who were resistant to or intolerant of hydroxyurea. We aimed to compare the long-term safety and efficacy of ruxolitinib with best available therapy in patients with polycythaemia vera who were resistant to or intolerant of hydroxyurea. Methods We report the 5-year results for a randomised, open-label, phase 3 study (RESPONSE) that enrolled patients at 109 sites across North America, South America, Europe, and the Asia-Pacific region. Patients (18 years or older) with polycythaemia vera who were resistant to or intolerant of hydroxyurea were randomly assigned 1:1 to receive either ruxolitinib or best available therapy. Patients randomly assigned to the ruxolitinib group received the drug orally at a starting dose of 10 mg twice a day. Single-agent best available therapy comprised hydroxyurea, interferon or pegylated interferon, pipobroman, anagrelide, approved immunomodulators, or observation without pharmacological treatment. The primary endpoint, composite response (patients who achieved both haematocrit control without phlebotomy and 35% or more reduction from baseline in spleen volume) at 32 weeeks was previously reported. Patients receiving best available therapy could cross over to ruxolitinib after week 32. We assessed the durability of primary composite response, complete haematological remission, overall clinicohaematological response, overall survival, patient-reported outcomes, and safety after 5-years of follow-up. This study is registered with ClinicalTrials.gov , NCT01243944 . Findings We enrolled patients between Oct 27, 2010, and Feb 13, 2013, and the study concluded on Feb 9, 2018. Of 342 individuals screened for eligibility, 222 patients were randomly assigned to receive ruxolitinib (n=110, 50%) or best available therapy (n=112, 50%). The median time since polycythaemia vera diagnosis was 8·2 years (IQR 3·9–12·3) in the ruxolitinib group and 9·3 years (4·9–13·8) in the best available therapy group. 98 (88%) of 112 patients initially randomly assigned to best available therapy crossed over to receive ruxolitinib and no patient remained on best available therapy after 80 weeks of study. Among 25 primary responders in the ruxolitinib group, six had progressed at the time of final analysis. At 5 years, the probability of maintaining primary composite response was 74% (95% CI 51–88). The probability of maintaining complete haematological remission was 55% (95% CI 32–73) and the probability of maintaining overall clinicohaematological responses was 67% (54–77). In the intention-to-treat analysis not accounting for crossover, the probability of survival at 5 years was 91·9% (84·4–95·9) with ruxolitinib therapy and 91·0% (82·8–95·4) with best available therapy. Anaemia was the most common adverse event in patients receiving ruxolitinib (rates per 100 patient-years of exposure were 8·9 for ruxolitinib and 8·8 for the crossover population), though most anaemia events were mild to moderate in severity (grade 1 or 2 anaemia rates per 100 patient-years of exposure were 8·0 for ruxolitinib and 8·2 for the crossover population). Non-haematological adverse events were generally lower with long-term ruxolitinib treatment than with best available therapy. Thromboembolic events were lower in the ruxolitinib group than the best available therapy group. There were two on-treatment deaths in the ruxolitinib group. One of these deaths was due to gastric adenocarcinoma, which was assessed by the investigator as related to ruxolitinib treatment. Interpretation We showed that ruxolitinib is a safe and effective long-term treatment option for patients with polycythaemia vera who are resistant to or intolerant of hydroxyurea. Taken together, ruxolitinib treatment offers the first widely approved therapeutic alternative for this post-hydroxyurea patient population. Funding Novartis Pharmaceuticals Corporation.

Background: Complete surgical resection is the standard treatment for patients with stage I-IIIA non-small cell lung cancer (NSCLC). Five-year survival rates range from 19% to 50%, with most patients dying from distant recurrence. Neoadjuvant or adjuvant chemotherapy improves overall survival (OS) by only 5% in patients with NSCLC, and new treatment options are needed. Preliminary data for PD1 or PD-L1 inhibitors as neoadjuvant therapy have shown major pathologic responses (MPRs) or pathologic complete responses (pCRs) in patients with early-stage NSCLC. The CANTOS study demonstrated reduced incidence of NSCLC and decreased lung cancer-related mortality with canakinumab (interleukin 1β inhibitor) versus placebo, in a dose-dependent manner for patients with atherosclerosis. In preclinical NSCLC humanized models, treatment with canakinumab ± anti-PD-1 inhibitor could lead to antitumor activity. Combination of canakinumab and pembrolizumab is expected to enhance the efficacy of PD-1 inhibition by inhibiting dysregulated inflammation in the tumor microenvironment. Based on available evidence, the CANOPY-N study was designed to evaluate the effect of canakinumab and pembrolizumab as monotherapy or in combination as neoadjuvant treatment for patients with resectable NSCLC. Methods: CANOPY-N (NCT03968419) is a Phase II, randomized, open-label study evaluating the effect of canakinumab and pembrolizumab as monotherapy or in combination as neoadjuvant treatment in patients with resectable NSCLC. Patients with histologically confirmed stage IB-IIIA NSCLC (excluding N2 and T4 tumors), no prior systemic therapy, Eastern Cooperative Oncology Group performance status 0 or 1, and eligibility for surgery and with a planned surgical resection in approximately 4-6 weeks (after first dose of study treatment) are eligible to participate. An archival (if obtained up to 6 months before first day of treatment) or new biopsy is required. Approximately 110 patients will be randomized in a 2:2:1 ratio (stratified by histology [squamous/nonsquamous]) to one of the treatment arms to receive a total of 2 doses (200 mg every 3 weeks) of canakinumab alone (n=44) or in combination with pembrolizumab (n=44) or pembrolizumab alone (n=22), with safety follow-up up to 130 days from last study drug dose. The primary endpoint is MPR rate (≤10% of residual viable tumor cells at time of surgery), and secondary endpoints include determination of overall response rate, MPR rate based on local review, surgical feasibility rates, incidence of antidrug antibodies, and pharmacokinetic parameters. As of October 19, 2020, there are 36 active study locations. Citation Format: Jay M. Lee, Tony Mok, Pilar Garrido, Edward S. Kim, Çağatay Arslan, Masahiro Tsuboi, Tuochuan Dong, Cecile Blin, Vanessa Rodrik-Outmezguine, Bijoyesh Mookerjee, Vanessa Passos, Jean-Louis Pujol. Canakinumab or pembrolizumab as monotherapy or in combination as neoadjuvant therapy in patients with surgically resected non-small cell lung cancer: CANOPY-N trial [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr CT237.

The sonidegib and ruxolitinib combination was assessed in an open-label study in JAK inhibitor-naive patients with myelofibrosis (MF). The primary objective of phase 1b was to establish the maximum tolerated dose (MTD) and/or recommended phase 2 dose (RP2D) and phase 2 was to assess spleen volume reduction at weeks 24 and 48. Fifty patients were enrolled. In the dose-escalation phase (n = 23), doses for sonidegib once daily/ruxolitinib twice daily were 400/10 mg (level 1, n = 8), 400/15 mg (level 2, n = 10), and 400/20 mg (level 3, n = 5). Two patients had dose-limiting toxicity at level 2: increased blood creatine phosphokinase (grades 3 and 4, n = 1 each). MTD/RP2D was determined as sonidegib 400 mg daily + ruxolitinib 20 mg twice daily. In phase 1b expansion and phase 2 stage 1 (n = 27), by weeks 24 and 48, ≥35% reduction in spleen volume was observed in 44.4% and 29.6% patients, respectively. By weeks 24 and 48, 42.0% and 26.0% patients had ≥50% reduction in Myelofibrosis Symptom Assessment Form total symptom score, respectively. Most common treatment-related adverse events (grade 3/4) were increased blood creatine phosphokinase (18%), anemia (14%), and thrombocytopenia (12%). Four deaths were reported due to multiple organ dysfunction syndrome (on-treatment; no relationship with study treatment), acute myeloid leukemia, MF progression, and aspiration pneumonia. Although well tolerated, this combination will not be further developed in MF patients due to modest overall benefit compared with historical ruxolitinib monotherapy. This trial was registered at www.clinicaltrials.gov as #NCT01787552.

Background/aim The phosphatidylinositol-3-kinase (PI3K) signaling pathway is frequently activated in cancer. Buparlisib (BKM120), an oral pan-PI3K inhibitor, inhibits proliferation of human cancer in preclinical models. Studies of buparlisib in Western and Japanese adults with advanced solid tumors established a recommended dose of 100 mg/day and showed an acceptable safety profile and evidence of efficacy. This phase I dose-escalation/expansion study aimed to establish the maximum tolerated dose (MTD) of single-agent, once daily oral buparlisib in Chinese patients with advanced solid tumors. Materials and methods Patients (n=32; primary tumor site: lung (n=15), breast (n=10) or head and neck (n=7); ≥2 prior lines of antineoplastic therapy (n=26)) received 80 mg (n=15) or 100 mg (n=17) daily buparlisib. Results Five patients experienced dose-limiting toxicities: grade (G)3 depression (n=1), G2 hyperglycemia (n=3) and G3 hyperglycemia (n=1). Most frequent buparlisib-related adverse events were hyperglycemia (n=18; 56%), alanine aminotransferase (ALT) and aspartate aminotransferase (AST) increase (n=9; 28%), as well as anxiety (n=6; 19%); most common buparlisib-related G3/4 adverse events: hyperglycemia (n=3; 9%), ALT and AST increase (n=2; 6%), as well as gamma-glutamyltransferase increase (n=2; 6%). Best response was stable disease (SD) in 10 patients (31%). Conclusion The MTD of buparlisib was declared as 100 mg/day. Safety, efficacy and pharmacokinetic data from this study were similar to those previously reported in Western and Japanese populations.

Myelofibrosis (MF) is associated with a variety of burdensome symptoms and reduced survival compared with age-/sex-matched controls. This analysis evaluated the long-term survival benefit with ruxolitinib, a Janus kinase (JAK)1/JAK2 inhibitor, in patients with intermediate-2 (int-2) or high-risk MF. This was an exploratory analysis of 5-year data pooled from the phase 3 COMFORT-I and -II trials. In both trials, patients could cross over to ruxolitinib from the control group (COMFORT-I, placebo; COMFORT-II, best available therapy). All continuing patients in the control groups crossed over to ruxolitinib by the 3-year follow-up. Overall survival (OS; a secondary endpoint in both trials) was evaluated using pooled intent-to-treat data from patients randomized to ruxolitinib or the control groups. OS was also evaluated in subgroups stratified by baseline anemia and transfusion status at week 24. A total of 528 patients were included in this analysis; 301 were originally randomized to ruxolitinib (COMFORT-I, n = 155; COMFORT-II, n = 146) and 227 to control (n = 154 and n = 73, respectively). The risk of death was reduced by 30% among patients randomized to ruxolitinib compared with patients in the control group (median OS, 5.3 vs 3.8 years, respectively; hazard ratio [HR], 0.70 [95% CI, 0.54–0.91]; P = 0.0065). After correcting for crossover using a rank-preserving structural failure time (RPSFT) method, the OS advantage was more pronounced for patients who were originally randomized to ruxolitinib compared with patients who crossed over from control to ruxolitinib (median OS, 5.3 vs 2.3 years; HR [ruxolitinib vs RPSFT], 0.35 [95% CI, 0.23–0.59]). An analysis of OS censoring patients at the time of crossover also demonstrated that ruxolitinib prolonged OS compared with control (median OS, 5.3 vs 2.4 years; HR [ruxolitinib vs censored at crossover], 0.53 [95% CI, 0.36–0.78]; P = 0.0013). The survival benefit with ruxolitinib was observed irrespective of baseline anemia status or transfusion requirements at week 24. These findings support ruxolitinib treatment for patients with int-2 or high-risk MF, regardless of anemia or transfusion status. Further analyses will be important for exploring ruxolitinib earlier in the disease course to assess the effect on the natural history of MF. ClinicalTrials.gov identifiers, NCT00952289 and NCT00934544 .

Introduction: The RESPONSE trial (NCT01243944) compared ruxolitinib (Rux) and best available therapy (BAT) in pts with polycythemia vera (PV) who were intolerant of or resistant to hydroxyurea (HU) according to modified European LeukemiaNET criteria. In the primary analysis, at week (Wk) 32, 60% of (pts) randomized to Rux achieved HCT control (HCT Methods: Adult pts with phlebotomy-dependent PV with splenomegaly, and resistant to or intolerant of HU were enrolled. Pts were randomized to receive Rux (at a starting dose of 10 mg BID) or single-agent BAT (1:1). HCT control was defined as lack of phlebotomy eligibility between Wks 8−32 with no more than 1 phlebotomy eligibility between randomization and Wk 8. Phlebotomy eligibility was based on protocol-defined HCT values (HCT > 45% and ≥ 3 percentage points higher than baseline or > 48%, whichever was lower; regardless of receipt of phlebotomy), and pts with missing data or assessments outside of protocol-defined time windows were considered non-responders. In this analysis, a logistic regression model was fitted to identify the significant baseline factors to predict HCT control response at Wk 32. Time to phlebotomy eligibility in the HCT control responders and time from the first phlebotomy eligibility to the second phlebotomy eligibility in the HCT control non-responders were plotted, and the changes in hematological parameters (HCT, WBC and platelet count), spleen volume and allele burden over time, up to Wk 256, were studied in HCT control responders and non-responders who were randomized to Rux treatment arm in RESPONSE. Results: A total of 222 pts were randomized to receive either Rux (n = 110) or BAT (n = 112). Baseline WBC (P=0.0198) and baseline JAK2 V617F allele burden (P=0.0159), were found to be predictors of the HCT response within Rux treated pt group (n = 110). In the HCT responder subgroup of the Rux arm, 23% (15/66) pts needed their first phlebotomy by Wk 256. In the HCT non-responder subgroup of the Rux arm, out of 28 patients who experienced their first phlebotomy between Wk 8 and Wk 32, 64% (18/28) of pts required subsequent phlebotomy by Wk 256, with a median duration of 28.4 Wks (12.7, NA). Pts receiving Rux demonstrated controlled hematologic parameters (HCT, WBC, and platelets) over the course of study, regardless of whether they were HCT control responders and HCT control non-responders at Wk 32. From Wk 48 to Wk 80, 97% HCT control responder pts and 84% HCT control non-responder pts of the Rux treatment arm required no phlebotomies. From Wk 80 to Wk 256, 91% and 68% of the evaluable pts in the Rux treatment arm remained phlebotomy-free for HCT control responders and non-responders, respectively. By Wk 256, spleen volume on an average was reduced from baseline by approximately 35% and 50% for HCT control responders and non-responders, respectively. In pts with available assessments, allele burden on an average was reduced approximately from 80% at baseline to 55% at Wk 256 in the HCT control responders, and approximately from 70% at baseline to 40% at Wk 256 in the HCT control non-responders. Conclusions: The results from present analysis demonstrated that the benefits of the Rux treatment were not limited to pts who achieved HCT control at Wk 32. Patients treated with Rux were able to maintain hematological parameters, spleen volume reduction, and JAK2 V617F allele burden reduction for a longer duration (up to 5 years), regardless of whether they were HCT control responders or non-responders at Wk 32. Disclosures Verstovsek: Constellation: Consultancy; Pragmatist: Consultancy; Incyte: Research Funding; Roche: Research Funding; NS Pharma: Research Funding; Celgene: Consultancy, Research Funding; Gilead: Research Funding; Promedior: Research Funding; CTI BioPharma Corp: Research Funding; Genetech: Research Funding; Blueprint Medicines Corp: Research Funding; Novartis: Consultancy, Research Funding; Sierra Oncology: Research Funding; Pharma Essentia: Research Funding; Astrazeneca: Research Funding; Ital Pharma: Research Funding; Protaganist Therapeutics: Research Funding. Kiladjian:Novartis: Honoraria, Research Funding; Celgene: Consultancy; AOP Orphan: Honoraria, Research Funding. Wroclawska:Novartis Pharma AG: Employment. Dong:Novartis: Employment. Vannucchi:Celgene: Membership on an entity's Board of Directors or advisory committees; CTI BioPharma: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Novartis: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Incyte: Membership on an entity's Board of Directors or advisory committees; Italfarmaco: Membership on an entity's Board of Directors or advisory committees.

Although most of the statistical methods for diagnostic studies focus on disease processes with binary disease status, many diseases can be naturally classified into three ordinal diagnostic categories, that is normal, early stage, and fully diseased. For such diseases, the volume under the ROC surface (VUS) is the most commonly used index of diagnostic accuracy. Because the early disease stage is most likely the optimal time window for therapeutic intervention, the sensitivity to the early diseased stage has been suggested as another diagnostic measure. For the purpose of comparing the diagnostic abilities on early disease detection between two markers, it is of interest to estimate the confidence interval of the difference between sensitivities to the early diseased stage. In this paper, we present both parametric and non-parametric methods for this purpose. An extensive simulation study is carried out for a variety of settings for the purpose of evaluating and comparing the performance of the proposed methods. A real example of Alzheimer's disease (AD) is analyzed using the proposed approaches.

Background RESPONSE is a randomized (1:1), open-label, phase 3 study of rux vs BAT in pts with PV who are resistant/intolerant of HU. In RESPONSE, significantly more pts achieved the primary composite end point (hematocrit control [Hct] and a ≥ 35% reduction in spleen volume [SVR]) with rux compared to BAT (23% vs 1%; P ˂ .0001). The 4-year (y) follow-up confirmed that primary and clinicohematologic (CLHM) responses were sustained with the rux treatment, and the safety profile was consistent with previous reports. Here, we report the long-term efficacy and safety after 5 ys (256 weeks [wks]) of treatment (after last pt first treatment visit) in RESPONSE. Methods PV pts resistant/intolerant of HU (per modified ELN criteria) with splenomegaly requiring phlebotomy (PBT) for Hct control were randomized to rux 10 mg BID or BAT (selected based on investigator's choice). The primary response was a composite end point of achieving both Hct control without PBT (defined as no PBT eligibility between wks 8 to 32 with no more than 1 PBT eligibility from randomization to wk 8) through wk 32 and a ≥ 35% SVR by imaging at wk 32. Key secondary endpoints were durability of the primary response, overall CLHM (Hct control without PBT, platelet count ≤ 400 × 109/L, WBC count ≤ 10 × 109/L, SVR ≥35% by imaging), and long-term safety. Overall survival (OS) was an exploratory end point. Pts randomized to BAT could cross over (CO) to rux after wk 32. Results Pts were randomized to rux (n = 110) and BAT (n = 112). Baseline characteristics were mostly balanced between arms. At baseline, pts randomized to rux were reported to have longer prior exposure to HU compared to BAT (162.9 vs 145.6 wks) and higher frequency of prior nonmelanoma skin cancer (NMSC) or pre-cancerous skin conditions (10.9 vs 6.3%). No pt remained on randomized BAT treatment after wk 80 and median CO time was 34.7 wks. At study completion, 72 pts (66%) in the rux arm and 64 of the 98 pts (65%) who crossed over from BAT completed 5 ys of on-study treatment. Main reason for discontinuation in the BAT arm was lack of efficacy (89%). Adverse events (AEs; 15% and 16%), disease progression (11% and 9% ) and pt decision (6% and 6%) primarily led to treatment discontinuations in the rux arm and CO population, respectively. At final analysis, the KM estimated probability of maintaining primary response for 224 wks (starting from wk 32) in the rux arm was 0.74 (95% CI: 0.51, 0.88) (Figure 1); 6/25 primary responders had progressed by the time of study completion. The probability of maintaining CLHM response for 224 wks (starting from wk 32) was 0.67 (95% CI: 0.54, 0.77) with 21/70 pts who achieved CLHM at wk 32 had progressed. Median duration of primary and CLHM responses had not been reached. In the intent-to-treat analysis not accounting for CO, KM estimates for OS at 5 ys in the rux and BAT arms were 91.9% (95% CI: 84.4, 95.9) and 91.0% (95% CI: 82.8, 95.4), respectively (HR = 0.95 [95% CI: 0.38, 2.41]). There were 2 on-treatment deaths in the rux arm (adenocarcinoma gastric [n = 1, investigator suspected event to be related to study drug] and neoplasm malignant [n = 1, not related to study drug]). In the CO population, 4 pts had fatal AEs leading to 4 on-treatment deaths (not related to rux). No pt died while on BAT. With longer exposure of rux, there was no notable increase in rate of AEs. The most common AEs (exposure-adjusted rate ≥ 5) in the rux arm and CO population were anemia (8.9 and 8.8), pruritus (7.0 and 6.1), diarrhea (7.0 and 3.6), weight increased (6.1 and 4.2), headache (5.8 and 5.2), arthralgia (5.6 and 3.3), fatigue (5.1 and 3.9) and muscle spasms (5.1 and 3.3). Rates (per 100 pt-y) of thromboembolic events were lower in rux-treated pts (1.2) and the CO population (2.7) compared to BAT pts (8.2). The exposure adjusted rates of second malignancies were 7.0 (rux) vs 4.5 (CO) vs 4.1 (BAT). The exposure-adjusted rate of NMSC was 5.1 (rux) vs 2.7 (CO) vs 2.7 (BAT). Rates of transformation to myelofibrosis and acute myeloid leukeimia in the rux arm were 2.1 and 0.2 per 100 pt-y, 1.8 and 0.6 per 100 pt-y in CO population, and 1.4 and 0.0 per 100 pt-y in BAT arm, respectively. Conclusion In HU resistant/intolerant PV pts, clinical benefits of rux treatment (Hct control and CLHM) were durable with long-term therapy. Considering that the OS findings from this analysis are confounded by extensive CO, the observed HR from this analysis represents a conservative estimate of rux benefit. The long-term safety was consistent with previous findings. Disclosures Kiladjian: AOP Orphan: Membership on an entity's Board of Directors or advisory committees, Research Funding; Novartis: Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees. Hino:Otsuka Pharmaceutical Co., Ltd.: Research Funding; Novartis: Research Funding. Pane:Novartis: Research Funding, Speakers Bureau; BMS: Speakers Bureau; AMGEN: Speakers Bureau. Masszi:Pfizer: Consultancy; BMS: Consultancy; AbbVie: Consultancy; Janssen-Cilag: Consultancy; Novartis: Consultancy; Takeda: Consultancy. Harrison:Celgene: Consultancy, Honoraria, Speakers Bureau; Roche: Consultancy, Honoraria; CTI BioPharma: Consultancy, Honoraria; Gilead: Honoraria, Speakers Bureau; Novartis: Consultancy, Honoraria, Research Funding, Speakers Bureau. Mesa:CTI: Research Funding; Promedior: Research Funding; Celgene: Research Funding; Incyte: Research Funding; Gilead: Research Funding; Galena: Consultancy; Ariad: Consultancy; Novartis: Consultancy. Miller:Gilead: Honoraria, Research Funding; Incyte: Consultancy, Honoraria, Research Funding, Speakers Bureau; CTI: Research Funding; Novartis: Honoraria, Speakers Bureau. Passamonti:Janssen: Consultancy, Speakers Bureau; Celgene: Consultancy, Speakers Bureau; Novartis: Consultancy, Honoraria, Speakers Bureau; Roche: Consultancy. Durrant:Novartis: Honoraria, Speakers Bureau. Griesshammer:Novartis: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau. Kirito:Novartis Pharma KK: Honoraria. Besses:Novartis: Honoraria, Research Funding; Shire: Honoraria. Francillard:Novartis: Employment. Dong:Novartis: Employment, Equity Ownership. Wroclawska:Novartis: Employment, Equity Ownership. Verstovsek:Incyte: Consultancy; Italfarmaco: Membership on an entity's Board of Directors or advisory committees; Novartis: Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Celgene: Membership on an entity's Board of Directors or advisory committees.

Many disease processes can be divided into three stages: i.e. the non-diseased stage, the early diseased stage and the fully diseased stage. To assess the accuracy of diagnostic tests for such diseases, various summary indexes have been proposed, such as volume under the surface (VUS), partial volume under the surface (PVUS), and the sensitivity to the early diseased stage given specificity and the sensitivity to the fully diseased stage (P2). This paper focuses on confidence interval estimation for P2 based on empirical likelihood. Simulation studies are carried out to assess the performance of the new methods compared to the existing parametric and non-parametric ones. A real data set from Alzheimer’s Disease Neuroimaging Initiative (ANDI)2 is analyzed. Key Words: Empirical Likelihood; Diagnostic tests; The sensitivity to the early diseased stage.

Although most of the statistical methods for diagnostic studies focus on disease processes with binary disease status, many diseases can be naturally classified into three ordinal diagnostic categories, that is normal, early stage, and fully diseased. For such diseases, the volume under the ROC surface (VUS) is the most commonly used index of diagnostic accuracy. Because the early disease stage is most likely the optimal time window for therapeutic intervention, the sensitivity to the early diseased stage has been suggested as another diagnostic measure. For the purpose of comparing the diagnostic abilities on early disease detection between two markers, it is of interest to estimate the confidence interval of the difference between sensitivities to the early diseased stage. In this paper, we present both parametric and non-parametric methods for this purpose. An extensive simulation study is carried out for a variety of settings for the purpose of evaluating and comparing the performance of the proposed methods. A real example of Alzheimer’s disease (AD) is analyzed using the proposed approaches.

In practice, there exist many disease processes with three ordinal disease classes, that is, the non-diseased stage, the early disease stage, and the fully diseased stage. Because early disease stage is likely the best time window for treatment interventions, it is important to have diagnostic tests that have good diagnostic ability to discriminate the early disease stage from the other two stages. In this paper, we present both parametric and non-parametric approaches for confidence interval estimation of probability of detecting early disease stage given the true classification rates for non-diseased group and diseased group, namely, the specificity and the sensitivity to full disease. We analyze a data set on the clinical diagnosis of early-stage Alzheimer’s disease from the neuropsychological database at the Washington University Alzheimer’s Disease Research Center using the proposed approaches. Copyright © 2011 John Wiley & Sons, Ltd.