Multi-agent Protocol Research Articles

An edge-based augmented system approach for second-order nodal consensus

Many works on multi-agent consensus protocol designs rely on the full states of nodes or the node- observer-based procedures, under both of which the nodal information plays a key role in the consensus design. In this paper, we propose an edge-based augmented system approach for second-order nodal consensus, based on the edge loop transfer matrix of the closed-loop multi-agent system, which is defined by cutting the loop at the inputs of the edges. Each agent is allowed to be of single-output and additional poles and zeros are introduced into edges to construct the channel filter with previous data stored in the memories. We firstly focus on the consensus design by reserving two terms of the non-recursive filter and then propose a simplified design approach exploiting a single parameter, which is of low complexity and can be readily implemented. We also discuss the consensus design by using the non-recursive channel filter with higher-order realizations. Finally, as an extension, we discuss the consensus design based on the first-order recursive channel filter. The effectiveness of the proposed method is verified by the illustrative simulations.

Pharmacologic Inhibition of NT5C2 Reverses Genetic and Nongenetic Drivers of 6-MP Resistance in Acute Lymphoblastic Leukemia.

Relapse-associated NT5C2 mutations directly contribute to relapse in ALL by driving resistance to chemotherapy with 6-MP. Pharmacologic inhibition of NT5C2 with CRCD2, a first-in-class nucleotidase inhibitor, enhances the cytotoxic effects of 6-MP and effectively reverses thiopurine resistance mediated by genetic and nongenetic mechanisms of NT5C2 activation in ALL. This article is highlighted in the In This Issue feature, p. 2483.

Optimal data transmission and pathfinding for WSN and decentralized IoT systems using I-GWO and Ex-GWO algorithms

Efficient resource use is a very important issue in wireless sensor networks and decentralized IoT-based systems. In this context, a smooth pathfinding mechanism can achieve this goal. However, since this problem is a Non-deterministic Polynomial-time (NP-hard) problem type, metaheuristic algorithms can be used. This article proposes two new energy-efficient routing methods based on Incremental Grey Wolf Optimization (I-GWO) and Expanded Grey Wolf Optimization (Ex-GWO) algorithms to find optimal paths. Moreover, in this study, a general architecture has been proposed, making it possible for many different metaheuristic algorithms to work in an adaptive manner as well as these algorithms. In the proposed methods, a new fitness function is defined to determine the next hop based on some parameters such as residual energy, traffic, distance, buffer size and hop size. These parameters are important measurements in subsequent node selections. The main purpose of these methods is to minimize traffic, improve fault tolerance in related systems, and increase reliability and lifetime. The two metaheuristic algorithms mentioned above are used to find the best values ​​for these parameters. The suggested methods find the best path of any length for the path between any source and destination node. In this study, no ready dataset was used, and the established network and system were run in the simulation environment. As a result, the optimal path has been discovered in terms of the minimum cost of the best paths obtained by the proposed methods. These methods can be very useful in decentralized peer-to-peer and distributed systems. The metrics for performance evaluation and comparisons are i) network lifetime, ii) the alive node ratio in the network, iii) the packet delivery ratio and lost data packets, iv) routing overhead, v) throughput, and vi) convergence behavior. According to the results, the proposed methods generally choose the most suitable and efficient ways with minimum cost. These methods are compared with Genetic Algorithm Based Routing (GAR), Artificial Bee Colony Based routing (ABCbased), Multi-Agent Protocol based on Ant Colony Optimization (MAP-ACO), and Wireless Sensor Networks based on Grey Wolf optimizer. (GWO-WSN) algorithms. The simulation results show that the proposed methods outperform the others.

Improvement and Optimization of a Mobile Multi-agent AODV Routing Protocol

<p>In the scenario of node failure in mobile multi-agent, Ad hoc On-demand Distance Vector (AODV) rout-ing protocol fails in network information transmission due to the damage of communication link. In addi-tion, the existing AODV multipath routing protocol node load balancing mechanism is imperfect. In order to solve the low network survivability caused by the above problems, an improved responsive multipath routing protocol MD_AODV (Multipath Destroy-resistance AODV) is proposed. The protocol is opti-mized in three aspects: multi-channel concurrency mechanism, node load balancing mechanism and shortest path maintenance mechanism. The simulation analysis of the modified AODV multipath routing protocol is realized on OPNENT simulation platform. The simulation results show that compared with AODV protocol, the improved multipath routing protocol can enhance the utilization of network re-sources and has better link redundancy. Moreover, the modified routing protocol not only enhances the reliability and fault tolerance of the path, but also increases the network aggregation bandwidth. </p> <p> </p>

Quantum and Classical Bayesian Agents

We describe a general approach to modeling rational decision-making agents who adopt either quantum or classical mechanics based on the Quantum Bayesian (QBist) approach to quantum theory. With the additional ingredient of a scheme by which the properties of one agent may influence another, we arrive at a flexible framework for treating multiple interacting quantum and classical Bayesian agents. We present simulations in several settings to illustrate our construction: quantum and classical agents receiving signals from an exogenous source, two interacting classical agents, two interacting quantum agents, and interactions between classical and quantum agents. A consistent treatment of multiple interacting users of quantum theory may allow us to properly interpret existing multi-agent protocols and could suggest new approaches in other areas such as quantum algorithm design.

Presentation, diagnosis, treatment and outcome of primary gastric lymphoma in 13 cats

Primary gastric lymphoma is a well-characterised disease in humans, but is poorly characterised in cats. This study retrospectively describes the presentation, diagnosis, treatment and outcome of cats with primary gastric lymphoma in a UK referral hospital. Medical records of cats diagnosed with primary gastric lymphoma, without ultrasonographic involvement of the intestinal tract, between 2009 and 2020 were reviewed. A total of 13 cats met the inclusion criteria. All cases were of large cell lymphoma. Cytology alone was diagnostic in nearly all cases (10/11). At diagnosis six cats were euthanised. The remaining seven cats were treated with a multiagent chemotherapy protocol (5/7) or a combination of prednisolone and chlorambucil (2/7). Median overall survival time was 300 days (ranging from 30–1980 days). The two cats treated with prednisolone and chlorambucil survived for 300 and 480 days respectively. This study raises awareness of feline primary gastric lymphoma for veterinary surgeons in clinical practice. Although an uncommon disease presentation, primary gastric lymphoma has unique characteristics that may differ from the high-grade intestinal form. Further studies are needed to evaluate the optimum therapeutic approach.

​Successful Medical Management of Lymphoma in Six Dogs

Background: Lymphoma is considered to be the most common hematopoietic tumor in dogs (83%) having an overall cure rate of less than five percent. Although many chemotherapy protocols have been suggested, doxorubicin based combination protocols are widely associated with long disease free interval and improved duration of survival. Six dogs presented to Madras Veterinary College Teaching Hospital with clinical signs suggestive of lymphoma were taken up for this study. Methods: The dogs were subjected to detailed clinical examination and dermatological examination. Fine needle aspiration cytology (FNAC) samples were taken from the lymph nodes in cases of multicentric lymphoma and from cutaneous nodules in cases of cutaneous lymphoma. Tru-cut biopsy and impression smears were collected from the cutaneous nodules. The FNAC and impression smears revealed pleomorphic lymphocytes suggestive of lymphoma. Ultrasonographic examination of all dog revealed the enlargement of mesenteric lymph nodes and spotted appearance of spleen. Treatment with multi-agent chemotherapeutic protocol (University of Wisconsin - Madison, Lymphoma CHOP - 19 protocol) was initiated after confirmative diagnosis. Result: Remissions in clinical signs were noticed within a week of therapy and no evidence of peripheral lymph node enlargement could be appreciated after three weeks of therapy. The diagnostic and management plans were presented and discussed.

THE CLOUD GNSS DATA FUSION APPROACH BASED ON THE MULTI-AGENT AUTHENTICATION PROTOCOLS’ ANALYSIS IN THE CORPORATE LOGISTICS MANAGEMENT SYSTEMS

Context. Modern corporate logistics management systems or tracking systems consist of wireless positioning systems. Typically, mobile technologies use signal receivers of satellites of GNSS. However, there is the general problem of data transmitting issues to analytic centers for further in the corporate logistics management systems. The object of the study was to develop the solutions alternative to GNSS.
 Objective. The goal of the work is the increasing vehicle location access control accuracy, based on the multi-agent authentication protocols’ analysis.
 Method. The study proposes the cloud data fusion platform offers to collect and archive data about all movable objects that can be on the road between departments inside the same warehouse area. Since there are operations with different wireless base stations' participation, so, the movable object can don’t enough time for a movement trajectory analysis in real-time. Thus, data processing must fulfill in the cloud dispatching center and data flow fusion is needed. An equations system that identifies the vehicles' location based on the method of multi-agent authentication process analysis after the GPS signal loss, was proposed. The multi-sectional configuration of the recursive neural network and the usage of cloud data fusion made it possible to increase the accuracy of vehicle location determination.
 Results. The developed method has been implemented in software and investigated for solving the problems of vehicle location control accuracy in the corporate logistics management systems.
 Conclusions. The conducted experiments have confirmed the proposed approach and allow recommending it for use in practice for solving the problems of improving the efficiency of vehicle location determination via the role-based access control integrated with multi-agent authentication service.

Approaches to identifying drug resistance mechanisms to clinically relevant treatments in childhood rhabdomyosarcoma.

Aim: Despite aggressive multiagent protocols, patients with metastatic rhabdomyosarcoma (RMS) have poor prognosis. In a recent high-risk trial (ARST0431), 25% of patients failed within the first year, while on therapy and 80% had tumor progression within 24 months. However, the mechanisms for tumor resistance are essentially unknown. Here we explore the use of preclinical models to develop resistance to complex chemotherapy regimens used in ARST0431. Methods: A Single Mouse Testing (SMT) protocol was used to evaluate the sensitivity of 34 RMS xenograft models to one cycle of vincristine, actinomycin D, cyclophosphamide (VAC) treatment. Tumor response was determined by caliper measurement, and tumor regression and event-free survival (EFS) were used as endpoints for evaluation. Treated tumors at regrowth were transplanted into recipient mice, and the treatment was repeated until tumors progressed during the treatment period (i.e., became resistant). At transplant, tumor tissue was stored for biochemical and omics analysis. Results: The sensitivity to VAC of 34 RMS models was determined. EFS varied from 3 weeks to > 20 weeks. Tumor models were classified as having intrinsic resistance, intermediate sensitivity, or high sensitivity to VAC therapy. Resistance to VAC was developed in multiple models after 2-5 cycles of therapy; however, there were examples where sensitivity remained unchanged after 3 cycles of treatment. Conclusion: The SMT approach allows for in vivo assessment of drug sensitivity and development of drug resistance in a large number of RMS models. As such, it provides a platform for assessing in vivo drug resistance mechanisms at a “population” level, simulating conditions in vivo that lead to clinical resistance. These VAC-resistant models represent “high-risk” tumors that mimic a preclinical phase 2 population and will be valuable for identifying novel agents active against VAC-resistant disease.

Towards a Formal Validation of a Hybrid Multi-Agent Routing Protocol for Ad hoc Networks

In the authors' previous papers, a new ant routing protocol for ad-hoc networks inspired from ant colony optimization was presented. They introduced a new approach that decreases both node energy consumption and routing overhead within the network. The validation of the routing protocol was based on series of simulations. The results show that the new algorithm provides a significant improvement compared to other protocols. After the algorithm is defined and published, they found it important to validate formally each one of its components in order to avoid any conflict, lack, or misbehaving situations. This process requires in a first step a formal specification. This is the main concern in this paper where the authors propose in a first part a formal specification using inference systems based on logical rules. A formal validation using these inference systems is proposed in a second step in order to prove the correctness, the soundness, the completeness, and the optimality of the proposition.

Are severe adverse events commonly observed in dogs during cancer chemotherapy? A retrospective study on 155 dogs.

Overall prevalence of severe adverse events (sAE) has been poorly studied in veterinary medicine and peer-reviewed studies mostly focused on a single protocol, making it difficult to have a general overview. The aim of this retrospective study was to assess the frequency and risk factors of sAE secondary to various protocols of chemotherapy in dogs. Medical records of 155 dogs receiving chemotherapy between January 2013 and December 2018 were reviewed. Adverse events (AE) were graded according to Veterinary Comparative Oncology Group-common terminology criteria for AE (VCOG-CTCAE) grading system. Statistical analyses were performed to determine whether demographic, cancer type and chemotherapy protocol were associated with development of sAE and their consequences. AE were reported at least once in 124 (80%) dogs and sAE were observed in 50 (32.3%) dogs. Among them, 23 (14.8%) had gastro-intestinal and 31 (20.0%) had myelotoxic events. sAE led to hospitalisation in 37 (23.9%) dogs, to chemotherapy arrest in 12 (7.7%) dogs and to euthanasia or death in 9 (5.8%) dogs. Haematopoietic tumours were statistically associated with a higher frequency of sAE (p=.004), gastrointestinal sAE (p=.009) and hospitalisation (p=.004). A body weight over 10kg was associated with less haematological sAE (p < .001). The use of a multi-agent protocol was highlighted as a risk factor for sAE (p=.038) and haematological sAE (p < .001). sAE following chemotherapy and leading to hospitalisation, chemo arrest or death were relatively common. A special attention during chemotherapy follow-up should be given to small dogs and those receiving multi-agent protocol or treated for haematopoietic tumours.

Network topology identification under the multi-agent agreement protocol

This paper investigates the problem of identifying the interaction geometry of a set of agents, whose collective goal are to achieve consensus under an agreement protocol. By classifying agents into different subsets based on their behavior, as well as introducing the so-called input and output agents, a relationship between the transfer function matrix and the identifiability of system parameters is established. Specifically, two cases are considered. If the set of input agents coincides with the set of output agents, the number of edges in the input agent set, in the complement of input agent set, and between these two sets can be uniquely identified. Thus, the search space of feasible graphs becomes much smaller. The problem can be solved in polynomial time, and an algorithm is provided. Moreover, if all the agents in the system are output agents, parameters of the system can be uniquely identified, and an algebraic method is given to exactly recover the graph topology. A numerical example illustrates the effectiveness of the proposed algorithm.

Prognostic factors of survival in patients with Burkitt lymphoma treated at the National Institute of Neoplastic Diseases in Peru: A 14-years’ experience.

e19570 Background: Burkitt lymphoma (BL) is one of the most aggressive B-cell lymphoma types. However, it has a high cure rate which varies depending on the different clinical pathological characteristics. We aimed to evaluate the prognostic factors for survival in patients with BL. Methods: We performed a retrospective study at the National Institute of Neoplastic Diseases in Peru from 2000 to 2014. We reviewed medical records of patients with BL aged 15 years and above. The data collected were based on clinical characteristics, treatment and survival. Patients with human immunodeficiency virus (HIV) infection were considered for analysis. The overall survival (OS) and progression-free survival (PFS) rates were estimated using the Kaplan-Meier method, the survival curves according to the clinical characteristics were compared using the log-rank test. Prognostic factors were determined using Cox regression modeling. Results: A total of 55 patients had sufficient data for analysis. The median age was 34 years (range 15-73), 78% were male and 73% had B symptoms. Lymph node involvement (LNI) as primary site was present in 89% of patients, of which 43% and 39% were retroperitoneal and cervical LNI, respectively. Bulky mass ( &gt; 10cm) was reported in 73% of the patients. Advanced disease was found in 78% (stage III 31%, IV 47%). Elevated serum LDH level was seen in 60% of cases; and 33% had HIV infection. Intensive chemotherapy was given to 80% (44) of patients, with 34% receiving multi-agent acute lymphoblastic leukemia protocol, 22.8% DA-EPOCH, 18% CODOX-M IVAC, and 11.4% Hyper-CVAD. No patient received rituximab in the front-line setting. The complete response rate was 39%. With a median follow-up of 9 years, the median OS for the entire cohort (N = 55) was 6.9 months (range 1.9-11.9) with a 5-year survival rate of 26.3%. The median OS for the treated patients (N = 44) was 11.3 months (range 3.2-19.4) with a 5-year survival rate of 30.6%. The median PFS was 7.3 months, and the 5-year OFS rate was 30.8%. 5-year OS rates were worse in those with B symptoms (66% vs. 12.5%, p = 0.003), elevated serum LDH level (44.1% vs. 15.2%, p = 0.043), albumin &lt; 3.5 g/dL (36.5% vs. 15.4%, p = 0.023), beta-2 microglobulin &gt; 2.5 mg/L (41.4% vs. 20%, p = 0.029), and HIV infection (31% vs. 16.7%, p = 0.006). In the multivariate analysis, the presence of B symptoms (HR 5.4 [2.2, 13.4], p &lt; 0.001) and HIV infection (HR 2.8 [1.5, 5.4], p = 0.002) were both factors associated with worse survival. Conclusions: In our study, the 5-year OS rate of patients with BL treated with chemotherapy was shorter (30.6%) than in developed countries. The response rates were also lower than in other studies. Inclusion of rituximab is needed to improve response rates. Presence of B symptoms and HIV infection were associated to worse survival.

Dec-SGTS: Decentralized Sub-Goal Tree Search for Multi-Agent Coordination

Multi-agent coordination tends to benefit from efficient communication, where cooperation often happens based on exchanging information about what the agents intend to do, i.e. intention sharing. It becomes a key problem to model the intention by some proper abstraction. Currently, it is either too coarse such as final goals or too fined as primitive steps, which is inefficient due to the lack of modularity and semantics. In this paper, we design a novel multi-agent coordination protocol based on subgoal intentions, defined as the probability distribution over feasible subgoal sequences. The subgoal intentions encode macro-action behaviors with modularity so as to facilitate joint decision making at higher abstraction. Built over the proposed protocol, we present Dec-SGTS (Decentralized Sub-Goal Tree Search) to solve decentralized online multi-agent planning hierarchically and efficiently. Each agent runs Dec-SGTS asynchronously by iteratively performing three phases including local sub-goal tree search, local subgoal intention update and global subgoal intention sharing. We conduct the experiments on courier dispatching problem, and the results show that Dec-SGTS achieves much better reward while enjoying a significant reduction of planning time and communication cost compared with Dec-MCTS (Decentralized Monte Carlo Tree Search).

Clinical Utility of Pegaspargase in Children, Adolescents and Young Adult Patients with Acute Lymphoblastic Leukemia: A Review.

Acute lymphoblastic leukemia (ALL) is a heterogenous hematological malignancy representing 25% of all cancers in children less than 15 years of age. Significant improvements in survival and cure rates have been made over the past four decades in pediatric ALL treatment. Asparaginases, derived from Escherichia coli and Erwinia chrysanthemi, have become a critical component of ALL therapy since the 1960s. Asparaginases cause depletion of serum asparagine, leading to deprivation of this critical amino acid for protein synthesis, and hence limit survival of lymphoblasts. Pegaspargase, a conjugate of monomethoxypolyethylene glycol (mPEG) and L-asparaginase, has become an integral component of pediatric upfront and relapsed ALL protocols due to its longer half-life and improved immunogenicity profile compared to native asparaginase preparations. Over the past two decades great strides have been made in outcomes for pediatric ALL due to risk stratification, incorporation of multiagent chemotherapy protocols, and central nervous system prophylaxis with pegaspargase having played an important role in this success. However, adolescents and young adults (AYA) with ALL when treated on contemporaneous trials using adult ALL regimens, continue to have poor outcomes. There is increasing realization of adapting pediatric trial regimens for treating AYAs, especially those incorporating higher intensity of chemotherapeutic agents with pegaspargase being one such agent. Dose or treatment-limiting toxicity is observed in 25–30% of patients, most notable being hypersensitivity reactions. Other toxicities include asparaginase-associated pancreatitis, thrombosis, liver dysfunction, osteonecrosis, and dyslipidemia. Discontinuation or subtherapeutic levels of asparaginase are associated with inferior disease-free survival leading to higher risk of relapse, and in cases of relapse, a higher risk for remission failure. This article provides an overview of available evidence for use of pegaspargase in pediatric acute lymphoblastic leukemia.

Formal modelling and verification of a multi-agent negotiation approach for airline operations control

This paper proposes and evaluates a new airline disruption management strategy using multi-agent system modelling, simulation, and verification. This new strategy is based on a multi-agent negotiation protocol and is compared with three airline strategies based on established industry practices. The application concerns Airline Operations Control whose core functionality is disruption management. To evaluate the new strategy, a rule-based multi-agent system model of the AOC and crew processes has been developed. This model is used to assess the effects of multi-agent negotiation on airline performance in the context of a challenging disruption scenario. For the specific scenario considered, the multi-agent negotiation strategy outperforms the established strategies when the agents involved in the negotiation are experts. Another important contribution is that the paper presents a logic-based ontology used for formal modelling and analysis of AOC workflows.

Hybrid-order Network Consensus for Distributed Multi-agent Systems

As an important field of Distributed artificial intelligence (DAI), multi-agent systems (MASs) have attracted the attention of extensive research scholars. Consensus as the most important issue in MAS, much progress has been made in studying the consensus control of MAS, but there are some problems remained largely unaddressed which cause the MAS to lose some useful network structure information. First, multi-agent consensus protocol usually proceeds over the low-order structure by only considering the direct edges between agents, but ignores the higher-order structure of the whole topology network. Second, the existing work assumes all the edges in a topology network have the same weight without exploring the potential diversity of the connections. In this way, multi-agent systems fail to enforce consensus, resulting in fragmentation into multiple clusters. To address the above issues, this paper proposes a Motif-aware Weighted Multi-agent System (MWMS) method for consensus control. We focus more on triangle motif in the network, but it can be extended to other kinds of motifs as well. First, a novel weighted network is used which is the combination of the edge-based lower-order structure and the motif-based higher-order structure, i.e., hybrid-order structure. Subsequently, by simultaneously considering the quantity and the quality of the connections in the network, a novel consensus framework for MAS is designed to update agents. Then, two baseline consensus algorithms are used in MWMS. In our experiments, we use ten topologies of different shapes, densities and ranges to comprehensively analyze the performance of our proposed algorithms. The simulation results show that the hybrid higher-order network can effectively enhance the consensus of the multi-agent system in different network topologies.

Resilient distributed economic dispatch of a cyber-power system under DoS attack

The economic dispatch problem of a smart grid under vicious denial of service (DoS) is the main focus of this paper. Taking the actual situation of power generation as a starting point, a new distributed optimization model is established which takes the environmental pollution penalty into account. For saving the limited bandwidth, a novel distributed event-triggered scheme is proposed to keep the resilience and economy of a class of cyber-power systems when the communication network is subject to malicious DoS attack. Then an improved multi-agent consensus protocol based on the gradient descent idea is designed to solve the minimization problem, and the prerequisites to minimize the system power generation cost are analyzed from the aspects of optimality and stability. Finally, the theoretical results are verified through a single-area 10-generator unit simulation.

Implementation of a multi-agent multi-criteria negotiation protocol for self-sustainable smart grids

ABSTRACT Smart grids, IoT and big data are the drivers of change in the energy sector. The presented research builds on these concepts by establishing a cooperative model that supports autonomous, self-sustainable and self-organised energy communities. It is based on a multi-agent and multi-criteria negotiation protocol in which consumers are involved on one side, while suppliers take part on the other side. The model maximises utility functions and gains of both sides according to the changing real-time demand and supply of electricity. The negotiation support system aims to automate, aid and structure the decision-making process in the energy sector by the means of monotonic concession protocol and autonomous agents. In addition, machine learning is applied to find patterns in big data pertaining to energy consumption, while blockchain is introduced to reliably store distributed data and to manage contractual relationships between actors in the market. The system hence consists of several interconnected modules that deal with data, knowledge, intelligence, decision-making, negotiation and communication.

Risk factors associated with progressive increases in serum creatinine concentrations in cats with cancer receiving doxorubicin.

BackgroundAzotemia occurs in cats administered doxorubicin, but risk factors have not been explored.ObjectiveTo determine incidence of progressive increases in serum creatinine concentration in cats with cancer receiving doxorubicin in single or multiagent chemotherapy protocols and associated risk factors.AnimalsSeventy cats with cancer receiving doxorubicin.MethodsA retrospective study (2007‐2017) of cats with indices of kidney function recorded before and after doxorubicin administration was reviewed. Cats diagnosed with kidney injury because of known etiologies other than possible doxorubicin toxicosis were excluded. Variables were compared to identify risk factors.ResultsMean age (±SD) was 10.9 years (±3.2). Cancer types included lymphoma (n = 36), sarcoma (n = 19) and carcinoma (n = 14). Chronic kidney disease was present in 29/70 (41%) cats before receiving doxorubicin. Of 70 cats, 24 (34%) developed an increase in serum creatinine concentration ≥0.3 mg/dL and 10 (14%) had an increase ≥50% from baseline. Mean time to increases in serum creatinine concentration ≥0.3 mg/dL from first administration of doxorubicin was 119.3 days (±89.7), with mean 2.8 (±1.2) doses administered. Neutropenia or anemia during chemotherapy and number of radiation therapy treatments under general anesthesia were risk factors for increases in serum creatinine concentration (P < .05). Cats receiving single agent doxorubicin had a higher likelihood of an increase in serum creatinine concentration ≥0.3 mg/dL from baseline than cats receiving CHOP‐based chemotherapy protocols (OR 20.0, 95% CI 2.9‐100).Conclusions and Clinical ImportanceProgressive increases in serum creatinine concentration from baseline were common in cats receiving doxorubicin and associated risk factors were identified.