Nonparametric analysis of delayed treatment effects using single-crossing constraints.

Clinical trials involving novel immuno-oncology therapies frequently exhibit survival profiles which violate the proportional hazards assumption due to a delay in treatment effect, and, in such settings, the survival curves in the two treatment arms may have a crossing before the two curves eventually separate. To flexibly model such scenarios, we describe a nonparametric approach for estimating the treatment arm-specific survival functions which constrains these two survival functions to cross at most once without making any additional assumptions about how the survival curves are related. A main advantage of our approach is that it provides an estimate of a crossing time if such a crossing exists, and, moreover, our method generates interpretable measures of treatment benefit including crossing-conditional survival probabilities and crossing-conditional estimates of restricted residual mean life. Our estimates of these measures may be used together with efficacy measures from a primary analysis to provide further insight into differences in survival across treatment arms. We demonstrate the use and effectiveness of our approach with a large simulation study and an analysis of reconstructed outcomes from a recent combination therapy trial.

Pembrolizumab monotherapy for the treatment of high-risk non-muscle-invasive bladder cancer unresponsive to BCG (KEYNOTE-057): an open-label, single-arm, multicentre, phase 2 study.

BACKGROUND: Standard treatment for high-risk non-muscle-invasive bladder cancer is transurethral resection of bladder tumour followed by intravesical BCG immunotherapy. However, despite high initial responses rates, up to 50% of patients have recurrence or become BCG-unresponsive. PD-1 pathway activation is implicated in BCG resistance. In the KEYNOTE-057 study, we evaluated pembrolizumab, a PD-1 inhibitor, in BCG-unresponsive non-muscle-invasive bladder cancer. METHODS: We did this open-label, single-arm, multicentre, phase 2 study in 54 sites (hospitals and cancer centres) in 14 countries. In cohort A of the trial, adults aged 18 years or older with histologically confirmed BCG-unresponsive carcinoma in situ of the bladder, with or without papillary tumours, with an Eastern Cooperative Oncology Group performance status of 0-2, and who were ineligible for or declined radical cystectomy were enrolled. All enrolled patients were assigned to receive pembrolizumab 200 mg intravenously every 3 weeks for up to 24 months or until centrally confirmed disease persistence, recurrence, or progression; unacceptable toxic effects; or withdrawal of consent. The primary endpoint was clinical complete response rate (absence of high-risk non-muscle-invasive bladder cancer or progressive disease), assessed by cystoscopy and urine cytology approximately 3 months after the first dose of study drug. Patient follow-ups were done every 3 months for the first 2 years and every 6 months thereafter for up to 5 years. Efficacy was assessed in all patients who received at least one dose of the study drug and met BCG-unresponsive criteria. Safety was assessed in all patients who received at least one dose of the study drug. This trial is registered with ClinicalTrials.gov number, NCT02625961, and is ongoing. FINDINGS: Between Dec 9, 2015, and April 1, 2018, we screened 334 patients for inclusion. 186 patients did not meet inclusion criteria, and 47 patients were assigned to cohort B (patients with BCG-unresponsive high grade Ta or any grade T1 papillary disease without carcinoma in situ; results will be reported separately). 101 eligible patients were enrolled and assigned to receive pembrolizumab. All 101 patients received at least one dose of the study drug and were included in the safety analysis. Five patients had disease that did not meet the US Food and Drug Administration definition of BCG-unresponsive non-muscle-invasive bladder cancer and were therefore not included in the efficacy analysis (n=96). Median follow-up was 36·4 months (IQR 32·0-40·7). 39 (41%; 95% CI 30·7-51·1) of 96 patients with BCG-unresponsive carcinoma in situ of the bladder with or without papillary tumours had a complete response at 3 months. Grade 3 or 4 treatment-related adverse events occurred in 13 (13%) patients; the most common were arthralgia (in two [2%] patients) and hyponatraemia (in three [3%] patients). Serious treatment-related adverse events occurred in eight (8%) patients. There were no deaths that were considered treatment related. INTERPRETATION: Pembrolizumab monotherapy was tolerable and showed promising antitumour activity in patients with BCG-unresponsive non-muscle-invasive bladder cancer who declined or were ineligible for radical cystectomy and should be considered a a clinically active non-surgical treatment option in this difficult-to-treat population. FUNDING: Merck Sharp & Dohme.

Pembrolizumab alone or combined with chemotherapy versus chemotherapy as first-line therapy for advanced urothelial carcinoma (KEYNOTE-361): a randomised, open-label, phase 3 trial.

BACKGROUND: PD-1 and PD-L1 inhibitors are active in metastatic urothelial carcinoma, but positive randomised data supporting their use as a first-line treatment are lacking. In this study we assessed outcomes with first-line pembrolizumab alone or combined with chemotherapy versus chemotherapy for patients with previously untreated advanced urothelial carcinoma. METHODS: KEYNOTE-361 is a randomised, open-label, phase 3 trial of patients aged at least 18 years, with untreated, locally advanced, unresectable, or metastatic urothelial carcinoma, with an Eastern Cooperative Oncology Group performance status of up to 2. Eligible patients were enrolled from 201 medical centres in 21 countries and randomly allocated (1:1:1) via an interactive voice-web response system to intravenous pembrolizumab 200 mg every 3 weeks for a maximum of 35 cycles plus intravenous chemotherapy (gemcitabine [1000 mg/m2] on days 1 and 8 and investigator's choice of cisplatin [70 mg/m2] or carboplatin [area under the curve 5] on day 1 of every 3-week cycle) for a maximum of six cycles, pembrolizumab alone, or chemotherapy alone, stratified by choice of platinum therapy and PD-L1 combined positive score (CPS). Neither patients nor investigators were masked to the treatment assignment or CPS. At protocol-specified final analysis, sequential hypothesis testing began with superiority of pembrolizumab plus chemotherapy versus chemotherapy alone in the total population (all patients randomly allocated to a treatment) for the dual primary endpoints of progression-free survival (p value boundary 0·0019), assessed by masked, independent central review, and overall survival (p value boundary 0·0142), followed by non-inferiority and superiority of overall survival for pembrolizumab versus chemotherapy in the patient population with CPS of at least 10 and in the total population (also a primary endpoint). Safety was assessed in the as-treated population (all patients who received at least one dose of study treatment). This study is completed and is no longer enrolling patients, and is registered at ClinicalTrials.gov, number NCT02853305. FINDINGS: Between Oct 19, 2016 and June 29, 2018, 1010 patients were enrolled and allocated to receive pembrolizumab plus chemotherapy (n=351), pembrolizumab monotherapy (n=307), or chemotherapy alone (n=352). Median follow-up was 31·7 months (IQR 27·7-36·0). Pembrolizumab plus chemotherapy versus chemotherapy did not significantly improve progression-free survival, with a median progression-free survival of 8·3 months (95% CI 7·5-8·5) in the pembrolizumab plus chemotherapy group versus 7·1 months (6·4-7·9) in the chemotherapy group (hazard ratio [HR] 0·78, 95% CI 0·65-0·93; p=0·0033), or overall survival, with a median overall survival of 17·0 months (14·5-19·5) in the pembrolizumab plus chemotherapy group versus 14·3 months (12·3-16·7) in the chemotherapy group (0·86, 0·72-1·02; p=0·0407). No further formal statistical hypothesis testing was done. In analyses of overall survival with pembrolizumab versus chemotherapy (now exploratory based on hierarchical statistical testing), overall survival was similar between these treatment groups, both in the total population (15·6 months [95% CI 12·1-17·9] with pembrolizumab vs 14·3 months [12·3-16·7] with chemotherapy; HR 0·92, 95% CI 0·77-1·11) and the population with CPS of at least 10 (16·1 months [13·6-19·9] with pembrolizumab vs 15·2 months [11·6-23·3] with chemotherapy; 1·01, 0·77-1·32). The most common grade 3 or 4 adverse event attributed to study treatment was anaemia with pembrolizumab plus chemotherapy (104 [30%] of 349 patients) or chemotherapy alone (112 [33%] of 342 patients), and diarrhoea, fatigue, and hyponatraemia (each affecting four [1%] of 302 patients) with pembrolizumab alone. Six (1%) of 1010 patients died due to an adverse event attributed to study treatment; two patients in each treatment group. One each occurred due to cardiac arrest and device-related sepsis in the pembrolizumab plus chemotherapy group, one each due to cardiac failure and malignant neoplasm progression in the pembrolizumab group, and one each due to myocardial infarction and ischaemic colitis in the chemotherapy group. INTERPRETATION: The addition of pembrolizumab to first-line platinum-based chemotherapy did not significantly improve efficacy and should not be widely adopted for treatment of advanced urothelial carcinoma. FUNDING: Merck Sharp and Dohme, a subsidiary of Merck, Kenilworth, NJ, USA.

KEYNOTE-676: Phase III study of BCG and pembrolizumab for persistent/recurrent high-risk NMIBC.

Background: Nonmuscle-invasive bladder cancer (NMIBC) is the most common form of bladder cancer, with high rates of disease recurrence and progression. Current treatment for high-risk NMIBC involves Bacillus Calmette-Guérin (BCG) therapy, but treatment options are limited for patients with recurrent or BCG-unresponsive disease. Aberrant programmed death 1 signaling has been implicated in BCG resistance and bladder cancer recurrence and progression, and pembrolizumab has shown efficacy in patients with BCG-unresponsive high-risk NMIBC. Aim: To describe the rationale and design for the randomized, comparator-controlled Phase III KEYNOTE-676 study, which will evaluate the efficacy and safety of pembrolizumab in combination with BCG in patients with persistent/recurrent high-risk NMIBC after BCG induction therapy. Trial registration number: NCT03711032.

Logistic Regression Likelihood Ratio Test Analysis for Detecting Signals of Adverse Events in Post-market Safety Surveillance.

The Vaccine Adverse Event Reporting System (VAERS) and other product surveillance systems compile reports of product-associated adverse events (AEs), and these reports may include a wide range of information including age, gender, and concomitant vaccines. Controlling for possible confounding variables such as these is an important task when utilizing surveillance systems to monitor post-market product safety. A common method for handling possible confounders is to compare observed product-AE combinations with adjusted baseline frequencies where the adjustments are made by stratifying on observable characteristics. Though approaches such as these have proven to be useful, in this article we propose a more flexible logistic regression approach which allows for covariates of all types rather than relying solely on stratification. Indeed, a main advantage of our approach is that the general regression framework provides flexibility to incorporate additional information such as demographic factors and concomitant vaccines. As part of our covariate-adjusted method, we outline a procedure for signal detection that accounts for multiple comparisons and controls the overall Type 1 error rate. To demonstrate the effectiveness of our approach, we illustrate our method with an example involving febrile convulsion, and we further evaluate its performance in a series of simulation studies.