Early Stopping Rules Research Articles

A hybrid deep recurrent artificial neural network with a simple exponential smoothing feedback mechanism

Both classical forecasting methods and machine learning approaches are used to solve forecasting problems. Deep artificial neural networks, one of the machine learning methods, are widely used today and give very good results. Recurrent neural networks, a type of deep neural network, are very important in forecasting problems. Simple recurrent artificial neural networks, which are the simplest deep recurrent neural networks, are often preferred in solving forecasting problems due to the small number of parameters they use. Simple exponential smoothing, one of the classical forecasting methods, attracts attention with its performance in solving forecasting problems. The motivation of the study is to create a new forecasting method by combining a classical and simple forecasting method with a deep recurrent artificial neural network in an architecture. In this, a new hybrid deep recurrent artificial neural network with a simple exponential smoothing feedback mechanism is proposed. The architecture of the proposed method is created as a combination of simple recurrent artificial neural networks and simple exponential smoothing methods. In the training of the proposed method, two training algorithms based on sine cosine optimization and particle swarm optimization algorithms are proposed. In these training algorithms, two different solution strategies such as restarting, and early stopping rule are used to avoid overfitting and local optimum problems. The performance of the proposed method is analysed using stock market datasets and compared with both different deep and shallow artificial neural networks and classical forecasting methods. As a result of the analyses, it is concluded that the proposed method is successful in one step ahead of forecasting performance.

Cross-validated risk scores adaptive enrichment (CADEN) design

We propose a Cross-validated ADaptive ENrichment design (CADEN) in which a trial population is enriched with a subpopulation of patients who are predicted to benefit from the treatment more than an average patient (the sensitive group). This subpopulation is found using a risk score constructed from the baseline (potentially high-dimensional) information about patients. The design incorporates an early stopping rule for futility. Simulation studies are used to assess the properties of CADEN against the original (non-enrichment) cross-validated risk scores (CVRS) design which constructs a risk score at the end of the trial. We show that when there exists a sensitive group of patients, CADEN achieves a higher power and a reduction in the expected sample size compared to the CVRS design. We illustrate the application of the design in two real clinical trials. We conclude that the new design offers improved statistical efficiency over the existing non-enrichment method, as well as increased benefit to patients. The method has been implemented in an R package caden.

Bayesian Interim Analysis and Efficiency of Phase III Randomized Trials.

Improving the efficiency of interim assessments in phase III trials should reduce trial costs, hasten the approval of efficacious therapies, and mitigate patient exposure to disadvantageous randomizations. We hypothesized that in silico Bayesian early stopping rules improve the efficiency of phase III trials compared with the original frequentist analysis without compromising overall interpretation. Cross-sectional analysis. 230 randomized phase III oncology trials enrolling 184,752 participants. Individual patient-level data were manually reconstructed from primary endpoint Kaplan-Meier curves. Trial accruals were simulated 100 times per trial and leveraged published patient outcomes such that only the accrual dynamics, and not the patient outcomes, were randomly varied. Early stopping was triggered per simulation if interim analysis demonstrated ≥ 85% probability of minimum clinically important difference/3 for efficacy or futility. Trial-level early closure was defined by stopping frequencies ≥ 0.75. A total of 12,451 simulations (54%) met early stopping criteria. Trial-level early stopping frequency was highly predictive of the published outcome (OR, 7.24; posterior probability of association, >99.99%; AUC, 0.91; P < 0.0001). Trial-level early closure was recommended for 82 trials (36%), including 62 trials (76%) which had performed frequentist interim analysis. Bayesian early stopping rules were 96% sensitive (95% CI, 91% to 98%) for detecting trials with a primary endpoint difference, and there was a high level of agreement in overall trial interpretation (Bayesian Cohen's κ, 0.95; 95% CrI, 0.92 to 0.99). However, Bayesian interim analysis was associated with >99.99% posterior probability of reducing patient enrollment requirements ( P < 0.0001), with an estimated cumulative enrollment reduction of 20,543 patients (11%; 89 patients averaged equally over all studied trials) and an estimated cumulative cost savings of 851 million USD (3.7 million USD averaged equally over all studied trials). Bayesian interim analyses may improve randomized trial efficiency by reducing enrollment requirements without compromising trial interpretation. Increased utilization of Bayesian interim analysis has the potential to reduce costs of late-phase trials, reduce patient exposures to ineffective therapies, and accelerate approvals of effective therapies. Question: What are the effects of Bayesian early stopping rules on the efficiency of phase III randomized oncology trials?Findings: Individual-patient level outcomes were reconstructed for 184,752 patients from 230 trials. Compared with the original interim analysis strategy, in silico Bayesian interim analysis reduced patient enrollment requirements and preserved the original trial interpretation. Meaning: Bayesian interim analysis may improve the efficiency of conducting randomized trials, leading to reduced costs, reduced exposure of patients to disadvantageous treatments, and accelerated approval of efficacious therapies.

N9: Pilot study of novel shortened induction for high-risk neuroblastoma.

10052 Background: N9 induction for high-risk neuroblastoma was designed to limit non-hematological toxicity (especially ototoxicity) and to prevent relapse in the central nervous system (CNS). N9 builds on 2 novel regimens: high-dose cyclophosphamide-topotecan-vincristine (CTV) and ifosfamide-carboplatin-etoposide (ICE), each with good penetration across the blood-brain barrier. Methods: N9 is a pilot study (Clinicaltrials.gov.NCT04947501) to assess feasibility and safety. Secondary/exploratory objectives include response, collection of peripheral blood stem cells (PBSCs; ≥5x106/kg CD34(+) cells sufficient for ≥2 rescues), tumor resection, and assessment of central nervous system relapse. Early stopping rules centered on excessive delay in timing of chemotherapy. Eligibility criteria included age &gt;1-to-&lt;13 years; 1 prior chemotherapy cycle was allowed. CTCAE Version 5.0 and International Neuroblastoma Response Criteria are used.N9 comprises 4 cycles of chemotherapy. Cycles start after absolute neutrophil count is &gt;500/μL, platelets &gt;100,000/μL, and non-hematologic toxicities grade ≤2. Intervals of 21-28 days between start of cycles are foreseen. PBSC collection and surgery follow &gt;3 cycles. Cycles #1 and #4 (CTV): cyclophosphamide 70mg/kg/day, days 1-2, topotecan 2mg/m2/day, days 1-4, and vincristine 0.067mg/kg, day 1. Cycle #2 (ICE): ifosfamide 1500mg/m2/day, days 1-5, carboplatin 400mg/m2/day, days 1-2, and etoposide 100mg/m2/day, days 1-5. Cycle #3: cyclophosphamide (as in CTV) and 72-hr infusions of doxorubicin 75mg/m2 and vincristine 0.067mg/kg. Results: The target number of 15 patients were enrolled: 10/2021-9/2023; age 1.5–9.8 (median 3.3) years; 14 stage M, 1 stage L2; and 10 post-1 cycle of other chemotherapy. They completed N9 without undue toxicity: all cycles started on time, organ functions remained intact, 11/11 patients tested had no ototoxicity, and acute toxicities were typical for myelosuppression (including uncomplicated fever/neutropenia). Responses were complete (n=6), partial (n=5), and stable disease (n=4). The target number of PBSCs was collected in 12 patients (9-75, median 14 x106/kg CD34(+) cells), 4x106/kg in 2 patients, and pending in 1. All patients had gross total resections of the primary tumor. Post-N9, patients did not undergo transplant, but proceeded to immunotherapy (naxitamab) or chemoimmunotherapy (naxitamab+irinotecan-temozolomide). Of 9 with residual disease post-N9, 7 achieved CR (median of 5.6 months from study entry) and 1 achieved metabolic CR (16 months), though 3 subsequently relapsed (no CNS). Conclusions: N9 shows promise for reducing chemotherapy and long-term toxicity. Less chemotherapy without compromising survival is a realistic goal by virtue of the advances with anti-GD2 monoclonal antibodies which, in combination with GM-CSF+low-dose chemotherapy, are highly effective against chemo-resistant disease in bone/bone marrow. Clinical trial information: NCT04947501 .

Digitally-enabled, person-centred care (PCC) in allergen immunotherapy: An ARIA-EAACI Position Paper.

In rhinitis and asthma, several mHealth apps have been developed but only a few have been validated. However, these apps have a high potential for improving person-centred care (PCC), especially in allergen immunotherapy (AIT). They can provide support in AIT initiation by selecting the appropriate patient and allergen shared decision-making. They can also help in (i) the evaluation of (early) efficacy, (ii) early and late stopping rules and (iii) the evaluation of (carried-over) efficacy after cessation of the treatment course. Future perspectives have been formulated in the first report of a joint task force (TF)-Allergic Rhinitis and Its Impact on Asthma (ARIA) and the European Academy of Allergy and Clinical Immunology (EAACI)-on digital biomarkers. The TF on AIT now aims to (i) outline the potential of the clinical applications of mHealth solutions, (ii) express their current limitations, (iii) make proposals regarding further developments for both clinical practice and scientific purpose and (iv) suggest which of the tools might best comply with the purpose of digitally-enabled PCC in AIT.

On the convergence of gradient descent for robust functional linear regression

Functional data analysis offers a set of statistical methods concerned with extracting insights from intrinsically infinite-dimensional data and has attracted considerable amount of attentions in the past few decades. In this paper, we study robust functional linear regression model with a scalar response and a functional predictor in the framework of reproducing kernel Hilbert spaces. A gradient descent algorithm with early stopping is introduced to solve the corresponding empirical risk minimization problem associated with robust loss functions. By appropriately selecting the early stopping rule and the scaling parameter of the robust losses, the convergence of the proposed algorithm is established when the response variable is bounded or satisfies a moment condition. Explicit learning rates with respect to both estimation and prediction error are provided in terms of regularity of the regression function and eigenvalue decay rate of the integral operator induced by the reproducing kernel and covariance function.

Prediction of slowdown of the Atlantic Meridional Overturning Circulation in coupled model simulations

In coupled perturbed parameter ensemble (PPE) experiments or for development of a single coupled global climate model (GCM) in general, models can exhibit a slowdown in the Atlantic Meridional Overturning Circulation (AMOC) that can result in unrealistically reduced transport of heat and other tracers. Here we propose a method that researchers running PPE experiments can apply to their own PPE to diagnose what controls the AMOC strength in their model and make predictions thereof. As an example, using data from a 25-member coupled PPE experiment performed with HadGEM3-GC3.05, we found four predictors based on surface heat and freshwater fluxes in four critical regions from the initial decade of the spinup phase that could accurately predict the AMOC transport in the later stage of the experiment. The method, to our knowledge, is novel in that it separates the effects of the drivers of AMOC change from the effects of the changed AMOC. The identified drivers are shown to be physically credible in that the PPE members exhibiting AMOC weakening possess some combination of the following characteristics: warmer ocean in the North Atlantic Subpolar Gyre, fresher Arctic and Tropical North Atlantic Oceans and larger runoff from the Amazon and Orinoco Rivers. These characteristics were further traced to regional responses in atmosphere-only experiments. This study suggests promising potential for early stopping rules for parameter perturbations that could end up with an unrealistically weak AMOC, saving valuable computational resources. Some of the four drivers are likely to be relevant to other climate models so this study is of interest to model developers who do not have a PPE.

Estimation of the odds ratio from multi-stage randomized trials.

A multi-stage design for a randomized trial is to allow early termination of the study when the experimental arm is found to have low or high efficacy compared to the control during the study. In such a trial, an early stopping rule results in bias in the maximum likelihood estimator of the treatment effect. We consider multi-stage randomized trials on a dichotomous outcome, such as treatment response, and investigate the estimation of the odds ratio. Typically, randomized phase II cancer clinical trials have two-stage designs with small sample sizes, which makes the estimation of odds ratio more challenging. In this paper, we evaluate several existing estimation methods of odds ratio and propose bias-corrected estimators for randomized multi-stage trials, including randomized phase II cancer clinical trials. Through numerical studies, the proposed estimators are shown to have a smaller bias and a smaller mean squared error overall.

Safety and efficacy of RCP recombinant spike protein covid-19 vaccine compared to Sinopharm BBIBP: A phase III, non-inferiority trial

BackgroundWe conducted a phase III, non-inferiority trial comparing safety and efficacy of RCP recombinant spike protein Covid-19 vaccine to BBIBP (Sinopharm). MethodsAdult Iranian population received RCP or BBIBP in a randomized, double blind and an additional non-randomized open labeled trial arms. Eligible participants signed a written informed consent and received two intramuscular injections three weeks apart. In the randomized arm, an intranasal dose of vaccine or adjuvant-only preparation were given to the RCP and BBIBP recipients at day 51 respectively. Participants were actively followed for up to 4 months for safety and efficacy outcomes. Primary outcome was PCR + symptomatic Covid-19 disease two weeks after the second dose. The non-inferiority margin was 10% of reported BBIBP vaccine efficacy (HR = 1.36). ResultsWe recruited 23,110 participants (7224 in the randomized and 15,886 in the non-randomized arm). We observed 604 primary outcome events during 4 months of active follow-up including 121 and 133 in the randomized and 157 and 193 cases in the non-randomized arms among recipients of RCP and BBIBP respectively. Adjusted hazard ratios for the primary outcome in those receiving RCP compared with BBIBP interval were 0.91 (0.71–1.16) and 0.62 (0.49–0.77) in the randomized and non-randomized arms respectively. The upper boundary of 99.1% confidence interval of HR = 0.91 (0.67–1.22) remained below the margin of non-inferiority in the randomized arm after observing the early stopping rules using O'Brien Fleming method. ConclusionOur study showed that the RCP efficacy is non-inferior and its safety profile is comparable to the BBIBP.

Bayesian phase II adaptive randomization by jointly modeling efficacy and toxicity as time-to-event outcomes

ABSTRACT The main goals of Phase II trials are to identify the therapeutic efficacy of new treatments and continue monitoring all the possible adverse effects. In Phase II trials, it is important to develop an adaptive randomization (AR) procedure that takes into account both the efficacy and toxicity. In most existing articles, toxicity is modeled as a binary endpoint through an unobservable random effect (frailty) to link the efficacy and toxicity. However, this approach does not capture toxicity profiles that evolve over time. In this article, we propose a new Bayesian adaptive randomization (BAR) procedure using the covariate-adjusted efficacy-toxicity ratio (ETR) index, where efficacy and toxicity are jointly modelled as time-to-event (TTE) outcomes. Furthermore, we also propose early stopping rules for toxicity and futility such that inferior treatments can be dropped at earlier time of trial. Simulation results show that compared to the BAR procedures based solely on the efficacy and that based on TTE efficacy and binary toxicity outcomes, the proposed BAR procedure can better identify the difference in treatment toxicity such that it can assign more patients to the superior treatment arm under some scenarios.

Fast iterative regularization by reusing data

Abstract Discrete inverse problems correspond to solving a system of equations in a stable way with respect to noise in the data. A typical approach to select a meaningful solution is to introduce a regularizer. While for most applications the regularizer is convex, in many cases it is neither smooth nor strongly convex. In this paper, we propose and study two new iterative regularization methods, based on a primal-dual algorithm, to regularize inverse problems efficiently. Our analysis, in the noise free case, provides convergence rates for the Lagrangian and the feasibility gap. In the noisy case, it provides stability bounds and early stopping rules with theoretical guarantees. The main novelty of our work is the exploitation of some a priori knowledge about the solution set: we show that the linear equations determined by the data can be used more than once along the iterations. We discuss various approaches to reuse linear equations that are at the same time consistent with our assumptions and flexible in the implementation. Finally, we illustrate our theoretical findings with numerical simulations for robust sparse recovery and image reconstruction. We confirm the efficiency of the proposed regularization approaches, comparing the results with state-of-the-art methods.

The Betel Nut Intervention Trial (BENIT)-A Randomized Clinical Trial for Areca Nut and Betel Quid Cessation: Primary Outcomes.

Areca nut and betel quid (ANBQ) chewing is a widespread carcinogenic habit. The BENIT (ClinicalTrials-NCT02942745) is the first known randomized trial designed for ANBQ chewers. We compared the intensive behavioral treatment intervention condition (IC) with the control condition (CC) in the BENIT and included a 5-stage early stopping rule. We report the primary analysis at stage 3. English-literate adults in Guam and Saipan who self-identified as ANBQ chewers with tobacco were enrolled between August 2016 and August 2020. IC participants (n = 88) received five in-person sessions over 22 days and a brochure containing quitting advice. CC participants (n = 88) received only the brochure. Participants were assessed at baseline and on day 22 of follow-up. Self-reported chewing status at day 22 was determined by a composite of two survey items with disparate wording and response options for cross-verification. Cessation rates were 38.6% (IC) and 9.1% (CC). Proportional hazards regression revealed a p = 0.0058, which met the Stage 3 criteria for significance, and an estimated reduction in ANBQ chewing for IC compared to the CC of 71% (95% CI: 41%-88%). Robust self-reported intervention effects at day 22 suggest that intensive cessation programs such as BENIT should be further developed and implemented on a larger scale.

Formula omitted]-EGAN: [formula omitted]-Energy distance GAN with an early stopping rule

Generative adversarial networks (GANs) are popular tools for learning the distribution of real samples and generating new ones, where the qualities of the generated images and the degree of preserved variation are two major concerns currently. In view of this, we propose an α-EGAN with energy distance as the loss function of the generator, which is proven to be effective in mitigating mode collapse. Moreover, an early stopping rule is proposed in the frame of hypothesis testing to avoid the unhealthy competition between the generator and the discriminator, thus achieving a trade-off between image qualities and variations. As a byproduct, the energy distance under the Euclidean norm can serve as a novel metric for evaluating generated samples in GAN. Experiments are conducted on simulated manifold datasets, as well as real MNIST and CelebA face datasets, showing that the proposed α-EGAN outperforms several competitors in both training stability and image quality.

Interim results of a phase II trial of first line retifanlimab (R) plus gemcitabine and docetaxel (GD) in patients (pts) with advanced soft tissue sarcoma (STS).

11518 Background: GD is an alternative to doxorubicin in the front line setting in advanced STS. In the Phase I portion of this study, the PD-1 antibody R, plus GD, was generally well tolerated. We hypothesized that R+GD would be more effective than historical controls treated with GD alone. Methods: This is an ongoing open-label single-center study of R+GD in pts with treatment-naïve unresectable or metastatic high-grade STS. G (900 mg/m2) is administered on days 1 and 8 and D (75 mg/m2) on day 8, in 21-day cycles. R (375 mg flat dose) is administered on day 1 starting in cycle 2 and continued as monotherapy ‘maintenance’ after completion of 6 cycles of GD. Up to fifty pts can accrue into five histology-specific cohorts of 10 pts each. The primary endpoint is to estimate the progression-free survival (PFS) rate at 24 weeks. Secondary endpoints included safety, best overall response rate (RR) by RECIST 1.1, disease control rate (DCR), and duration of response (DOR). Subgroup analyses to evaluate all endpoints within each histology-specific cohort are preplanned. An early stopping rule for excessive toxicity will halt accrual if a prespecified severe adverse event rate is surpassed. Results: As of January 11, 2023, 43 pts were enrolled and treated with R+GD. The leiomyosarcoma (LMS), undifferentiated pleomorphic sarcoma (UPS)/myxofibrosarcoma (MFS), dedifferentiated liposarcoma (DDLPS), and other STS cohorts (n = 10 pts each) were fully accrued. Three of 10 angiosarcoma (AS) pts were enrolled. Median age was 59.5 (range 21 – 81) and 27 (63%) were male. Two DDLPS patients were not yet evaluable for response. Of 41 evaluable pts, the best overall RR was 22% (95% confidence interval [CI]) 11 – 38). Confirmed responses were seen in UPS/MFS (n = 4), LMS (2), AS (2), and follicular dendritic cell sarcoma (1). Three pts (LMS, ossifying fibromyxoid tumor, AS) had an unconfirmed PR, for a best RR of 29% (95% CI 16 – 46). Median PFS was 32.7 weeks (95% CI 26.4 – not estimable [NE]) and median DOR was 24 weeks (95% CI 15 – NE). Safety was evaluated in all 43 pts and the early stopping rule was not triggered. Eighteen (42%) pts had at least one Grade (Gr) 3 or 4 treatment-related adverse event (TRAE). The most common ( &gt; 5%) were anemia (16%), neutropenia (9%), febrile neutropenia (7%), lung infection (7%), and leukopenia (7%). Seven pts (16%) had pneumonitis: one Gr 1, four Gr 2, and two Gr 3. Six pts (14%) stopped treatment due to toxicity, including five with pneumonitis. Conclusions: The median PFS of R+GD appears promising compared to historical controls, although the primary endpoint analysis is pending completion of accrual. There was a higher incidence of pneumonitis with the combination compared to GD alone. Future studies of this combination will need to carefully consider the benefits and risks after the final efficacy and safety analyses are performed. Clinical trial information: NCT04577014 .

The PRIME-CUT study: A single-arm phase 2 study of ADT with PD-1 blockade and docetaxel in men with metastatic hormone-sensitive prostate cancer.

e17089 Background: Overall survival of patients (pts) with metastatic hormone-sensitive prostate cancer (mHSPC) is improved with androgen deprivation therapy (ADT) and treatment intensification. Initially, ADT induces a robust and functional immune infiltrate into the tumor microenvironment (TME), with upregulation of immune checkpoint molecules. These effects diminish as castration-resistant prostate cancer (CRPC) invariably develops. We report the first results of the phase 2 PRIME-CUT study (modulating the PRostate cancer Immune MicroEnvironment with ChemoimmUnoTherapy for metastatic prostate cancer). Methods: Eligible pts had newly diagnosed mHSPC, ECOG performance status (PS) 0 or 1, serum testosterone levels &gt; 150 ng/dL, agreed to undergo serial metastatic tumor biopsies, and had not received ADT in the prior 6 months. Treatment consisted of ADT (degarelix 240 mg subcutaneously (SC) for one dose, followed by leuprolide 22.5 mg SC every 12 weeks) followed by anti-PD-1 antibody (cemiplimab 350 mg IV every 3 weeks) beginning four weeks after ADT initiation. With the intent of immune priming, a two-cycle lead-in of anti-PD-1 therapy was administered prior to standard of care docetaxel (75 mg/m2 every 3 weeks for six cycles). Participants received ADT and cemiplimab until study completion (52 weeks) or until a lack of clinical benefit or intolerable side effects. The primary endpoint was the rate of undetectable PSA at 6 months after chemotherapy initiation (week 37) and compared to the historical rate from the phase III CHAARTED trial (32%). Subjects were monitored for toxicity using a Bayesian adaptive study design with an early stopping rule for toxicity. Results: The study completed accrual of 20 pts (median age 62 years) from Columbia University Irving Medical Center in Aug 17, 2022. Among them, 7 self-identified race as Black and 12 self-identified their ethnicity as Hispanic. Baseline PSA ranged from 8.15 to &gt; 5000 ng/dL (median 2513.1 ng/dL). Site of metastatic disease included: 17/20 bone, 12/20 lymph nodes, 11/20 visceral organs. Overall, 14 of 20 patients received at least 4 cycles of docetaxel. Of the 12 patients that were still on-study at 37 weeks, 2 had an undetectable PSA, and 4 additional pts had PSA &lt; 1.0 ng/dL. Twelve pts experienced ≥ Grade 3 adverse events (AEs), 6 AEs were attributed to study intervention including cytopenias (3), hypotension (1), pneumonitis (1), and stenosis of gastrointestinal stoma (1). Unfortunately, two patients deceased on-study (myocardial infarction after 13 weeks, severe SARS-CoV-2 infection after 19 weeks). Conclusions: The Prime-Cut study did not meet its primary endpoint. However, PSA kinetics and results from the pre-planned correlative science demonstrate ongoing role for immunotherapeutic approaches in mHSPC. No new safety signals were observed. Clinical trial information: NCT03951831 .

A phase 1a/1b study of aurora kinase A inhibitor VIC-1911 as monotherapy and in combination with sotorasib for the treatment of KRAS G12C-mutant non–small-cell lung cancer.

TPS9140 Background: Activity of direct covalent KRASG12C inhibitors in NSCLC is limited by mutational and adaptive resistance. AURKA has roles in centrosome maturation and cytokinesis. AURKA amplification/overexpression is reported in multiple tumors, including NSCLC, and AURKA signaling may mediate adaptive resistance to KRAS inhibition. VIC-1911 is a highly selective, orally active small molecule inhibitor of AURKA. VIC-1911 demonstrated monotherapy activity in vitro in KRASG12C-mutant human NSCLC cells with intrinsic and acquired resistance to the KRASG12C inhibitor sotorasib, and combination VIC-1911 and sotorasib showed synergy in the same cell lines. Combination VIC-1911 and sotorasib delayed the emergence of adaptive resistance in vitro. These findings suggest that 1) AURKA activation is present in both intrinsic and acquired resistance to sotorasib in KRASG12C-mutant NSCLC and 2) the combination of VIC-1911 and sotorasib is a potential therapeutic approach for KRASG12C-mutant NSCLC with intrinsic or acquired resistance to sotorasib monotherapy. In addition, in vivo data suggest both sotorasib and adagrasib are synergistic in combination with VIC-1911 in human KRASG12C-mutant NSCLC cell line xenograft models. Methods: A non-randomized, open-label Phase 1a/1b study of VIC-1911, an aurora kinase A inhibitor, administered as monotherapy and in combination with sotorasib for the treatment of advanced KRASG12C-mutant NSCLC is in progress. Participants ≥18 years, with locally advanced or metastatic KRASG12C-mutant NSCLC, prior treatment with at least 1 line of PD-1/PD-L1 inhibitor therapy with or without platinum-based chemotherapy and have adequate organ function are eligible. Both KRASG12C inhibitor naïve and experienced patients are eligible. VIC-1911 monotherapy is given at oral doses of 25 – 90 mg twice daily in dose escalation and at the VIC-1911 maximum tolerated dose (MTD) in the expansion cohort in patients who are refractory/relapsed on prior KRASG12C inhibitor therapy. The combination regimen includes dose escalation with VIC-1911 at oral doses of 75, 150 or 200 mg twice daily on days 1-4, 8-11 and 15-18 every 28-day cycle plus sotorasib 960 mg daily in patients who are refractory/relapsed or naïve to prior KRASG12C inhibitor therapy. In dose escalation, a 3+3 schema is followed with dose limiting toxicity (DLT) criteria for toxicity and a sample size ≤ 36. For expansion cohorts, the sample size is based on Simon 2-stage optimal design, with a sample size ≤ 104. VIC-1911 and sotorasib dose modification criteria are provided. Supportive medications are allowed and early stopping rules for toxicity will be followed. VIC-1911 pharmacokinetics will be assessed. Pharmacodynamics will be determined from ctDNA and tumor biomarker analysis. Clinical trial information: NCT05374538 .

Modeling the impact of xenointoxication in dogs to halt Trypanosoma cruzi transmission.

Chagas disease, a vector-borne parasitic disease caused by Trypanosoma cruzi, affects millions in the Americas. Dogs are important reservoirs of the parasite. Under laboratory conditions, canine treatment with the systemic insecticide fluralaner demonstrated efficacy in killing Triatoma infestans and T. brasiliensis, T. cruzi vectors, when they feed on dogs. This form of pest control is called xenointoxication. However, T. cruzi can also be transmitted orally when mammals ingest infected bugs, so there is potential for dogs to become infected upon consuming infected bugs killed by the treatment. Xenointoxication thereby has two contrasting effects on dogs: decreasing the number of insects feeding on the dogs but increasing opportunities for exposure to T. cruzi via oral transmission to dogs ingesting infected insects. Examine the potential for increased infection rates of T. cruzi in dogs following xenointoxication. We built a deterministic mathematical model, based on the Ross-MacDonald malaria model, to investigate the net effect of fluralaner treatment on the prevalence of T. cruzi infection in dogs in different epidemiologic scenarios. We drew upon published data on the change in percentage of bugs killed that fed on treated dogs over days post treatment. Parameters were adjusted to mimic three scenarios of T. cruzi transmission: high and low disease prevalence and domestic vectors, and low disease prevalence and sylvatic vectors. In regions with high endemic disease prevalence in dogs and domestic vectors, prevalence of infected dogs initially increases but subsequently declines before eventually rising back to the initial equilibrium following one fluralaner treatment. In regions of low prevalence and domestic or sylvatic vectors, however, treatment seems to be detrimental. In these regions our models suggest a potential for a rise in dog prevalence, due to oral transmission from dead infected bugs. Xenointoxication could be a beneficial and novel One Health intervention in regions with high prevalence of T. cruzi and domestic vectors. In regions with low prevalence and domestic or sylvatic vectors, there is potential harm. Field trials should be carefully designed to closely follow treated dogs and include early stopping rules if incidence among treated dogs exceeds that of controls.

Stochastic mirror descent method for linear ill-posed problems in Banach spaces

Consider linear ill-posed problems governed by the system for , where each A i is a bounded linear operator from a Banach space X to a Hilbert space Y i . In case p is huge, solving the problem by an iterative regularization method using the whole information at each iteration step can be very expensive, due to the huge amount of memory and excessive computational load per iteration. To solve such large-scale ill-posed systems efficiently, we develop in this paper a stochastic mirror descent method which uses only a small portion of randomly selected equations at each iteration step and incorporates convex regularization terms into the algorithm design. Therefore, our method scales very well with the problem size and has the capability of capturing features of sought solutions. The convergence property of the method depends crucially on the choice of step-sizes. We consider various rules for choosing step-sizes and obtain convergence results under a priori early stopping rules. In particular, by incorporating the spirit of the discrepancy principle we propose a choice rule of step-sizes which can efficiently suppress the oscillations in iterates and reduce the effect of semi-convergence. Furthermore, we establish an order optimal convergence rate result when the sought solution satisfies a benchmark source condition. Various numerical simulations are reported to test the performance of the method.

Abstract OT2-15-01: COntinue the SaMe systemic therapy after local ablative therapy for Oligoprogression in metastatic breast cancer - the COSMO study

Abstract Background Patients with metastatic breast cancer (MBC) are usually treated with palliative systemic therapy until progression or unacceptable toxicity. When progression occurs, clinicians typically switch to the next line of systemic therapy. However, this strategy might be suboptimal in case of mixed response or oligoprogression. Oligoprogression refers to a situation in which inter-lesion heterogeneity causes one or few metastatic lesions to progress while the vast majority of the metastatic burden responds to the treatment. Locally ablative treatment of the progressive lesion(s) with radiotherapy, resection or radiofrequency ablation may enable continuation of otherwise effective systemic treatment. Solid data to support this approach, however, is lacking at this moment. Study design The COSMO study is an investigator-initiated single-arm phase 2 study. It is intended to be a multicenter study. Eligibility criteria Patients with MBC and oligoprogression, defined as 1-2 progressive lesions while the majority of the metastatic burden remains stable or in remission, are eligible. Current systemic therapy must be 1st or 2nd line and patients must have responded to this systemic therapy for at least six months prior to the occurence of oligoprogression. The aberrant lesion must be amenable to locally ablative therapy. All eligibility criteria may be found in table 1: eligibility criteria. Aim To investigate efficacy of local ablation of aberrant lesions combined with continuation of the same systemic therapy in patients with oligoprogression of MBC. Primary endpoint: Progression-free survival rate at six months (PFS-6). Secondary endpoints: - PFS - Overall survival - Time to next line of systemic therapy Primary and secondary endpoints will also be stratified by localization of progressive lesion and BC subtype (ER+/HER2- vs. HER2+ vs. TN). - Local control rate of lesion treated with LAT at 12 months - Complications due to LAT - Quality of life - Incidence of visceral crisis Exploratory endpoints: - Biomarkers, including receptor (ER/PR/HER2), PD-L1 and gene expression patterns in biopsies from progressive and responding lesions - Prognostic value of ctDNA at oligoprogression and during the course of treatment Statistical methods An A’hern design was chosen to conduct this study. As null-hypothesis (H0), PFS-6 rate ≤ 25% is set. The alternative hypothesis (H1) is PFS-6-rate ≥ 40%. Hypotheses were set to meet criteria for clinical significance, as stated by experts in the field. Setting one-sided α = 0.05 and a desirable power=95%, 107 evaluable patients are needed. To account for 10% of the patients to be unevaluable, 118 included patients are needed. An early stopping rule for safety is applicable: if out of the first 45 patients ≥4 of these patients have developed visceral crisis, the study will be closed early. Accrual Start: July 2022 Target: 118 patients Estimated accrual time: 2-3 years Further information Corresponding author: c.almekinders@nki.nl Clinicaltrials.gov identifier: NCT05301881 This study is funded by the Maarten van der Weijden foundation. This study was developed during the 21st EORTC/ESMO/AACR workshop on methods in clinical cancer research (MCCR). Table 1. Eligibility criteria. Citation Format: Cornelia AM Almekinders, Tessa G Steenbruggen, Ingrid A. Mandjes, Marta Lopez-Yurda, Monique EMM Bos, Terry G Wiersma, Gabe S. Sonke. COntinue the SaMe systemic therapy after local ablative therapy for Oligoprogression in metastatic breast cancer - the COSMO study [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr OT2-15-01.

Distributed Multi-View Sparse Vector Recovery

In this paper, we consider a multi-view compressed sensing problem, where each sensor can only obtain a partial view of the global sparse vector. Here the partial view means that some arbitrary and unknown indices of the global vector are unobservable to that sensor and do not contribute to the measurement outputs. The sensors aim to collaboratively recover the global state vector in a decentralized manner. We formulate this recovery problem as a bilinear optimization problem relying on a factored joint sparsity model (FJSM), in which the variables are factorized into a node-specific sparse local masking vector and the desired common sparse global vector. We first theoretically analyze the general conditions guaranteeing the global vector's successful recovery. Then we propose a novel in-network algorithm based on the powerful distributed alternating direction method of multipliers (ADMM), which can reconstruct the vectors and achieve consensus among nodes concerning the estimation of the global vector. Specifically, each node alternately updates the common global vector and its local masking vector, and then it transfers the estimated global vector to its neighboring nodes for further updates. To avoid potential divergence of the iterative algorithm, we propose an early stopping rule for the estimation of the local masking vectors and further conceive an estimation error-mitigation algorithm. The convergence of the proposed algorithms is theoretically proved. Finally, extensive simulations validate their excellent performance both in terms of the convergence and recovery accuracy.