Personalized medicine in hematologic malignancies: the ex-csepa model's role in Acinetobacter Baumannii bacteremiaAbstract
The Ex-CSEPA model was developed to predict 30-day mortality for patients with hematologic malignancies after a standard 14-day antibiotic (ATB) treatment for Acinetobacter baumannii bacteremia. This study systematically evaluated the model’s performance in stratifying patientsinto high-risk and low-risk categories and compared the efficacy of extended ATB therapy with the standard 14-day course. The Ex-CSEPA model integrates several critical factors: extended course ATB (Ex), clinical improvement (C), starting ATB time (S), ECOG score (E), Pitt’s bacteremia score (P), and APACHE II score (A) to comprehensively assess mortality risk. The findings demonstrated that the Ex-CSEPA model effectively distinguished between high-risk and low-risk patients, with significant differences in mortality risks. High-risk patients had lower median survival compared to low-risk patients (p-value < 0.001), confirming the model’s accuracy. While extended ATB therapy did not improve overall survival rates, high-risk individuals benefited significantly. Extended ATB therapy reduced mortality by 32% and increased survival time by 66%, according to multivariable and Weibull proportional-hazards regression analyses. This study emphasizes the advantage of extended therapy for high-risk individuals and the usefulness of the Ex-CSEPA model in directing personalized treatment. In conclusion, the Ex-CSEPA model is a robust tool with substantial clinical implications, supporting personalized interventions with extended ATB therapy for high-risk patients and validating the standard 14-day course for low-risk individuals.
Keywords:
Hematologic malignancy; Acinetobacter baumannii; Bloodstream infection; 30-day survival; Predictive model.
INTRODUCTION
Hematologic malignancies (HM) cause significant immune dysfunction and impair the production of normal blood cells, causing patients to be exceptionally vulnerable to life-threatening infections (Allegra et al., 2021; Chandra et al., 2022; Plackoska, Shaban, Nijnik, 2022). These malignancies disrupt both innate and adaptive immune pathways, with specific effects varying by HM type. For example, leukemia is associated with the overproduction of malignant leukocytes, which suppress normal hematopoiesis, leading to severe neutropenia and lymphopenia that hinder the immune system's ability to combat infections. In contrast, B-cell lymphomas often result in hypogammaglobulinemia, reducing immunoglobulin levels and increasing susceptibility to encapsulated bacterial infections. Chemotherapy further exacerbates these defects, inducing prolonged periods of immunosuppression and damaging mucosal barriers critical for infection defense.
These vulnerabilities are amplified by the risk of acquiring infections caused by multidrug-resistant (MDR) organisms, such as Acinetobacter baumannii (AB). HM patients are particularly predisposed to MDR infections due to their frequent exposure to healthcare settings, prolonged hospitalization, and extensive use of broad-spectrum antibiotics, which exert selective pressure favoring resistant strains (Baker, Satlin, 2016; Park et al., 2024). Moreover, their impaired immune defenses provide an ideal environment for MDR pathogens to proliferate unchecked, unlike in other immunocompromised populations where immune recovery may occur more rapidly. Among MDR pathogens, AB poses significant challenges due to its ability to form biofilms, evade host defenses, and exhibit resistance to multiple antibiotic classes, leading to limited treatment options and high mortality rates. The factors mentioned above exacerbate morbidity and mortality among affected individuals (Contejean et al., 2016). The presence of infection episodes, especially when accompanied by febrile neutropenia, underscores the severity of the situation, as 30-day mortality rates can increase by as much as 40% (Nørgaard et al., 2006). The presence of multidrug-resistant strains has further complicated the situation (Chaiwarit et al., 2005; Turkoglu et al. 2011; Garcia-Vidal et al., 2018), necessitating extended periods of antibiotic treatment despite ongoing discussions regarding its effectiveness and associated risk of complications. The controversy surrounding extended antibiotic treatment in immunocompromised patients infected with difficult-to-treat pathogens follows from the balance between the potential benefits of prolonged therapy in eradicating the infection and the associated risks of adverse effects, including antimicrobial resistance and drug toxicity (De Waele, Martin-Loeches, 2018; Sousa, et al., 2019; Tansarli et al., 2019). While prolonged antibiotic courses may prevent recurrence by ensuring complete pathogen eradication, they simultaneously elevate the risk of resistance emergence, limiting future therapeutic options (Katip, Uitrakul, Oberdorfer, 2021). Additionally, disruptions to the normal microbiota and drug-related toxicities increase the risk of adverse effects associated with protracted antibiotic use. These effects can predispose patients to develop secondary infections and other complications (Sousa et al., 2019; Tansarli et al., 2019). As a result, the optimal duration of antibiotic therapy must be determined through a comprehensive consideration of patient-specific factors, the characteristics of the underlying infection, and the current state of evidence concerning the effectiveness and safety of the medications.
To assist patients at this dangerous juncture, this study introduces "Ex-CSEPA”, a revolutionary mortality prediction tool. Designed with HM patients experiencing AB bloodstream infections while on a standard 14-day antibiotic regimen in mind, this predictive model boasts a remarkable sensitivity and accuracy rate of 98.7%. This model has the potential to significantly reduce mortality rates among this exceptionally vulnerable patient population by providing clinicians with actionable insights regarding patient risk stratification, specifically in identifying individuals who require prolonged antibiotic therapy. "Outcomes of 30-Day Survival in Hematologic Malignancy" analyzed in this study is the comparison between lowand high-risk cohorts of patients who have AB bloodstream infection, implementing the “Ex-CSEPA” predictive model.
The predictive model was systematically constructed by conducting a comprehensive analysis of five critical factors, which were carefully selected using an allencompassing methodology. The patient's Eastern Cooperative Oncology Group (ECOG) performance status (Azam et al., 2019), Pitt's bacteremia score (Vaquero- Herrero et al., 2017), Acute Physiology and Chronic Health Evaluation (APACHE) II score (Knaus et al., 1985), time-to-start AB-specific antibiotic (ATB) therapy (Im et al., 2022), and clinical response after a 14-day treatment regimen (Tong-Minh et al., 2021), were the factors considered. Patients were carefully categorized into specific risk groups: low and high, based on their cumulative scores calculated from these characteristics. The group classified as low risk, with scores between -14.5 and 0, had an exceptionally reduced likelihood of a fatality within 30 days, by approximately 1.67%. In contrast, the group classified as high-risk, with scores ranging from 0.5 to 12.5 points, had a significantly greater probability of mortality within the identical period, reaching a concerning rate of 99.07%. The validated Ex-CSEPA model scoring system is shown in Table I.
Theimplementationofthiscomprehensive risk stratification methodology enabled accurate detection of individuals who were at an increased mortality risk, allowing medical professionals to personalize treatment plans accordingly and potentially reduce adverse effects among this vulnerable group of patients. The purpose of this study is to investigate the 30-day survival rates of patients with HM who suffered from bloodstream infections caused by AB. A comparison is made between high-risk and low-risk cohorts using the Ex-CSEPA predictive model. The study assesses the efficacy of the Ex-CSEPA predictive model in determining between patient cohorts with low and high risk by comparing their rates of survival.
In addition to employing the Ex-CSEPA predictive model to stratify patients into high and low-risk groups, this study will investigate the efficacy of prolonged ATB therapy in the high-risk group. Extended antibiotic therapy involves the administration of antibiotics for a time that exceeds the typical 14-day treatment period. The objective of this strategy is to maximize the efficiency of systemic infection management in individuals diagnosed with hematologic malignancy (HM), who have a heightened susceptibility to negative consequences owing to their compromised immune systems and pre-existing medical conditions. The study aims to assess whether extended antibiotic therapy provides any survival advantage in high-risk patients by comparing the 30-day mortality rates between those in obtaining extended ATB therapy and those receiving standard ATB therapy. This investigation will further improve our comprehension of the effectiveness of prolonged antibiotic therapy in enhancing outcomes for patients with HM who have bloodstream infections caused by AB. Ultimately, this will provide valuable information for clinical practice and treatment guidelines.
MATERIAL AND METHODS
A retrospective cohort analysis was conducted from March 2019 to February 2024. The study included hospitalized patients who were admitted to a government hospital in Chiang Mai, Thailand. These individuals had been diagnosed with HM and had completed a 14- day ATB program to treat AB. This study was approved under Certificate No. NKP 079/65 by the Human Research Ethics Committee of Nakornping Hospital. It adhered to ethical guidelines including the Declaration of Helsinki, the Belmont Report, CIOMS guidelines, and ICH-GCP standards. Informed consent was waived for this retrospective cohort study due to its minimal risk and the utilization of pre-existing medical records. Stringent confidentiality measures were implemented to ensure full compliance with ethical principles throughout the study.
The criteria for inclusion in the study were as follows: (i) patients aged 18 years or older who were diagnosed with HM according to ICD10 codes C081 to C096 including lymphomas, leukemias, myelomas, and other related malignancies, and bacteremia according to ICD10 R78.81; (ii) patients with documented evidence of AB bacteremia; (iii) patients who had completed a 14-day course of ATB treatment for AB bacteremia; and (iv) patients who were admitted to the study site within the specified timeframe. The exclusion criteria consisted of (i) individuals with an Eastern Cooperative Oncology Group (ECOG) score of 4 or higher; and (ii) situations in which recovery from AB infection was achieved, but death resulted from illness caused by other pathogens.
The study cohort was classified into high-risk and low-risk groups based on the Ex-CSEPA model scoring system. This stratification method, described in Table I, categorized patients according to the risk of mortality within 30 days.
Outcomes Assessment
The primary outcome was mortality within 30 days of collecting AB bloodstream-infected specimens. Clinical response, and microbiological response were secondary outcomes. Clinical response was defined as complete or partial remission of infectious symptoms. Clinical failures went under all clinical response criteria. Two negative AB hemocultures examined the microbiological response after the initial positive hemoculture. Finally, acute nephrotoxicity was tested for AKIN-classified acute kidney damage (AKI) (Lopes, Jorge, 2013). The baseline renal function was determined using the latest blood creatinine measurement before to hospitalization or, when neither was available, the lowest recorded value during the hospital admission. Pre-existing kidney illness was incorporated into the analysis by including chronic kidney disease (CKD)as a variable in the Charlson comorbidity score and multivariable regression analysis.
Antibiotic Susceptible testing
The Nakornping Hospital Clinical Microbiology Department detected an AB bloodstream infection by traditional hemoculture and biochemistry. Aerobic and anaerobic growth conditions have been designed into standard bottles carrying rich media. Up to 10 mL of blood can be accommodated by them. The addition of lytic chemicals to certain growth media enhances the revival and proliferation of organisms that have been engulfed by phagocytes. The typical duration for incubation is 5 days, which is adequate for the retrieval of most organisms. Nevertheless, it is necessary to extend the incubation period for species with slow growth rates, such as fungi and Mycobacteria spp. Contemporary laboratories depend on automated incubators that incorporate constant monitoring to discover positive bottles. This greatly reduces the workload, incubation time, and rate of contamination. Blood-culture positive is often determined by monitoring the creation of CO2 by growing microorganisms, which leads to an increase in pH that can be observed through changes in color, fluorescence signal, or redox fluctuations. The susceptibility of AB to antibiotics was assessed using the VITEK 2 method, whereas the susceptibility to ATB was tested using broth microdilution. Amikacin, gentamicin, piperacillin/tazobactam, ceftazidime, ampicillin/sulbactam, meropenem, imipenem, ciprofloxacin, and colistin were tested for antibiotic susceptibility following the CLSI procedure (Weinstein, Lewis 2020).
Statistical Analysis
A descriptive analysis was conducted to evaluate the clinical relevance, prognostic factors, disease, treatment regimen, and participant characteristics. Categorical variables were summarized using frequencies and percentages, while continuous variables were summarized using means and standard deviations. The efficacy of the extended ATB course was determined by the odds ratio (OR) between groups using a multivariable risk regression model with a Poisson variance function. The multivariable regression model was modified by incorporating a composite prognostic factor comprising age, body mass index (BMI), ECOG performance status (Sørensen et al., 1993), ICU status, ventilator requirement, vasoconstrictor requirement, Pitt's bacteremia score (Rhee et al., 2009), APACHE II score (Knaus et al., 1985), and occurrence of acute nephrotoxicity. Furthermore, a propensity analysis was carried out in order to refine the model.
The propensity scores have been calculated using stabilized inverted-probability weighting, taking into account characteristics that indicate the likelihood of obtaining the lengthier 14-day antibiotic treatment course. The parameters considered in this study were age, BMI, ECOG performance status, absolute neutrophil count (ANC), ICU status, ventilator need, vasoconstrictor need, Charlson comorbidity index, Pitt's bacteremia score, APACHE II score, granulocytecolony stimulating growth factor (G-CSF) use, and acute nephrotoxicity.
The 30-day survival rates were further analyzed using Weibull proportional hazard regression, a full parametric method, to account for the timing of events. The comprehensive evaluation of survival probabilities was facilitated by this sophisticated statistical procedure, which considered both the overall patient cohort and specific high-risk subgroups. This approach enabled the evaluation of survival probability while accounting for the fluctuating hazard rates over time.
The statistical tests conducted were two-tailed, and the threshold for statistical significance was setat a p-value of less than 0.05. In order to differentiate between the two groups with respect to mortality, a sample size of 26 patients was deemed necessary through power analysis.
RESULTS
The trial included a cohort of 82 individuals who successfully finished a 14-day course of antibiotic treatment. Out of the total, 24 patients were categorized as high risk, whilst 58 were categorized as low risk. Among these patients, 50 individuals were administered long-term ATB treatment, while 32 individuals received short-term treatment (Table II). The majority of the study population consisted of males (68.3%), with an average age of 52 years. The most prevalent diagnoses were acute myeloid leukemia (AML) at 23.2%, diffuse large B-cell lymphoma (DLBCL) at 18.3%, and multiple myeloma (MM) at 14.6%. A large proportion of individuals had hematologic malignancies at an advanced stage or with a high risk, and they had a poor performance status, as indicated by an ECOG score of 2 or higher. Most of cases (93.9%) were diagnosed with infections caused by carbapenem-resistant Acinetobacter baumannii (CRAB). The microbiological success of all patients was confirmed, and 70.7% of patients exhibited a clinical response following the 14-day treatment course. However, 57.3% of patients developed acute renal damage following treatment. The Ex-CSEPA model was implemented to classify the patients into highand low-risk groups while conducting the analysis in the present research. The comparison revealed significant disparities between these groups in terms of a variety of prognostic factors, which reflected the corresponding mortality risks. Specifically, the group of individuals at high risk showed an assortment of characteristics that indicated a less favorable outcome compared to those at low risk. Significant factors that appeared more prevalent in the high-risk group included advanced age, higher ECOG performance status scores indicating poorer functional status, admission to the intensive care unit (ICU), requirement for mechanical ventilation, need for vasoactive agents to maintain hemodynamic stability, elevated Charlson comorbidity index scores reflecting greater comorbidity burden, higher Pitt's bacteremia scores indicative of more severe illness, and elevated APACHE II scores suggesting greater disease severity. In addition, the group at high risk encountered delayed in starting antibiotic therapy that were specific to AB-bacteremia, which could potentially affect the final outcome of the treatment. Additionally, acute kidney injury (AKI) was more prevalent in the highrisk group, particularly among individuals with reduced baseline glomerular filtration rates (GFR), suggesting that renal impairment has a compounding effect on mortality risk. These findings emphasize the complex interaction of several clinical factors in affecting mortality outcomes and emphasize the significance of risk stratification using Ex-CSEPA model for directing clinical care methods in these selected patients.
In order to evaluate the clinical advantages of prolonging the ATB course beyond 14 days, a study monitored the 30-day survival of patients from the inception of infection signs and symptoms. In the extended-antibiotic (ATB) group, the average duration for AB-specific antibiotic therapy was found to be 17 days, with a standard deviation (SD) of 0.4. Although the majority of prognostic factors were initially comparable between the standard and prolonged ATB groups, a distinct pattern appeared for patients with higher potential to receive the extended treatment. These individuals typically exhibited a higher age, elevated ECOG scores, lower absolute neutrophil counts (ANC), ICU admissions, ventilator and vasoactive agent requirements, and increased Charlson comorbidity indices, Pitt's bacteremia scores, APACHE II scores, receiving granulocyte-colony stimulating factor (G-CSF), and incidences of acute kidney injury (AKI). To address this tendency, a propensity score analysis was implemented utilizing the stabilized inverse-probability weighting method to adjust the cohort's balance prior to analyzing effectiveness. This statistical adjustment assured that the prognostic factors were comparable and similar between the groups following the modifications. The evaluation of the standard deviation of the means, as shown in Table III, verified an equitable allocation of prognostic factors, allowing for an accurate comparison of clinical outcomes related to the extended antibiotic treatment.
Prognostic factors of Extended-ATB vs 14-day ATB original cohort and post-propensity score adjustment cohort standard difference
The patient characteristics of individuals who received an extended antibiotic course and those who completed the standard 14-day regimen are compared, that has been stratified by risk subgroups. The results suggest that the prognostic factors of both high-risk and low-risk groups were mainly comparable. However, notable variations arose among individuals in the highrisk group who underwent extended antibiotic (ATB) treatment. Particularly, these patients were often older and evidenced lower absolute neutrophil counts (ANC). Additionally, the prolonged ATB regimen was unable to achieve a clinical response in all high-risk patients, suggesting an increased likelihood of worsening. The aforementioned observed differences, despite having significance, were adjusted using propensity score methodologies to ensure an equitable comparison between the treatment groups.
The Ex-CSEPA model demonstrated robust performance in stratifying individuals according to their risk for mortality over a 30-day observation period. Out of the 82 patients observed, 24 patients (29.3%) had passed away, all of whom were classified as high-risk by the Ex-CSEPA model. In stark contrast, all patients in the low-risk group survived the observation period. The two groups had very different median survival times. The high-risk group had a median survival time of 17 days, with an interquartile range (IQR) of 16 to 19 days, while the low-risk group had a median survival time of 30 days, with an IQR of 30 to 30 days. This difference was statistically significant (p-value < 0.001). This outcome underscores the model's efficacy in distinguishing patients with significantly different prognostic outcomes, thereby validating its utility in clinical risk stratification.
The efficacy of the prolonged course of ATB has been assessed by multivariable regression analysis, indicating no noteworthy disparity in the overall survival of all patients. However, a significant advantage of the extended ATB course was discovered when the Ex- CSEPA model score was used to determine the outcome. The multivariable analysis adjusted by composite prognostic factors revealed an odd ratio (OR) of 0.71 with a 95% confidence interval (CI) of 0.52-0.97 and a p-value of 0.034. The propensity score analysis further confirmed the consistent results. Stabilized inverseprobability weighting indicated an OR of 0.68, a 95% CI of 0.47-0.99, and a p-value of 0.045.
A comprehensive parametric analysis was conducted to further analyze the 30-day survival rates, specifically employing the Weibull proportional hazard regression model to account for the timing of events. This advanced analytical method confirmed the results of the regression study that had previously been performed. In cases where the entire patient population was taken into account, no substantial difference in clinical benefit was observed between the extended ATB course and the standard 14-day regimen. Nevertheless, the extended ATB course demonstrated a significant superiority in the high-risk patient group, emphasizing the potential advantages of extended therapy in this specific subgroup.
A multivariable analysis of Weibull proportional hazards regression was used to conduct a 30-day survival analysis comparing extended-ATB therapy to the normal 14-day ATB therapy, as demonstrated in Figure 1. The analysis demonstrated a hazard ratio (HR) of 0.41, with a 95% confidence interval (CI) of 0.17-0.98, and a p-value of 0.045. This indicates a statistically significant decrease in the risk of mortality when using extended-ATB therapy. After adjusting the analysis using the propensity score, the hazard ratio declined to 0.34, with a 95% confidence interval of 0.13-0.85, and a p-value of 0.021. This modification emphasizes the robustness of the results, indicating that extended-ATB therapy may provide a significant survival advantage, particularly in the scenario of high-risk patients.
(A) Weibull regression all individual and (B) specific high-risk groups stratified by extended-ATB and 14-day course.
DISCUSSION
The investigation comprised a cohort of 82 patients who successfully completed a 14-day antibiotic treatment regimen. The group of participants was stratified into high-risk and low-risk categories using the Ex-CSEPA model scoring system, resulting in 24 patients assigned to high-risk and 58 patients assigned to low-risk. Significant differences in prognostic factors that were associated with mortality risks between the two groups were revealed by the Ex-CSEPA model, which accordingly demonstrated the model's effectiveness. Survival analysis demonstrated that the high-risk group experienced a significantly shorter survival time than the low-risk group. The significant disparity in survival durations demonstrates the accuracy and reliability of the Ex-CSEPA model in estimating the likelihood of mortality within 30 days for patients with hematologic malignancies. The odds ratio (OR) and hazard ratio (HR) in our research offer distinct perspectives on the effect of prolonged antibiotic medication on survival in the high-risk cohort. The odds ratio of 0.71, obtained using multivariable logistic regression, indicates a 29% decrease in the probability of mortality associated with extended therapy, demonstrating its comprehensive protective impact. The HR of 0.41, derived from a Weibull regression model, signifies a 59% decrease in mortality risk over time, underscoring the enduring survival advantage of prolonged medication across the research duration. These findings indicate the clinical importance of extended antibiotic treatment in enhancing patient outcomes.
The potential to improve clinical decision-making is the major significance of this discovery. Clinicians can more effectively customize treatment strategies by precisely identifying high-risk patients. For individuals at greater risk, the results indicate that providing more intensive care, which might require extending the use of antibiotics beyond the usual 14-day treatment, could be advantageous. By focusing on the unique requirements of patients at higher risk of mortality, this tailored approach has the potential to enhance patient outcomes, optimize resource allocation, and improve the overall quality of care. The Ex-CSEPA model functions as both a prediction tool and an index for personalized therapy strategies in clinical practice.
This study investigated the effectiveness of extended-ATB therapy, primarily targeting patients with hematologic malignancy categorized into high-risk and low-risk subgroups based on the Ex-CSEPA model. Although the extended ATB therapy did not demonstrate a substantial difference in overall survival across the entire patient cohort, a significant benefit was observed in the high-risk group. The extending ATB treatment resulted in a significant augmentation in survival outcomes for patients at high risk. The multivariable regression investigation, after accounting for several prognostic indicators, consistently demonstrated that patients at high risk encountered substantial benefits from an extended duration of antibiotic treatment. The results were additionally supported by propensity score analysis, which further confirmed the survival benefit for this particular cohort. The extensive parametric analysis applying the Weibull proportional hazard regression model further confirmed these observations. Although the clinical benefit of extended versus standard ATB therapy was negligible for the entire population, the analysis revealed that high-risk patients experienced a substantial increase in survival rates when treated with the extended regimen. In the context of our investigation, extended-ATB therapy exhibited an important impact on mortality rates. Based on the multivariable regression analysis, the extended-ATB regimen contributed to a 32% reduction in mortality rates. In addition, the examination of survival time using the Weibull proportional-hazards regression model demonstrated a more significant advantage, with a 66% augmentation in survival time observed in individuals who received the extended-ATB treatment. The results highlight the significant improvement in survival rates associated with the extended-ATB therapy, especially for patients classified as high-risk according to the Ex- CSEPA model.
The research emphasizes the potential of extended-ATB therapy to considerably reduce the risks of fatalities in individuals at high risk, asdetermined by the Ex-CSEPA model. The personalized strategy of prolonged antibiotic therapy for individualswith a greater likelihood of negative consequences emphasizes the significance of individualized medical care. Clinicians could improve patient outcomes and optimize the consumption of medical resources by identifying and targeting high-risk patients with a more intensive treatment strategy. In clinical practice, the findings underscore the urgent necessity of precise risk stratification tools, such as the Ex-CSEPA model. These models empower healthcare practitioners to make well-informed judgments regarding the duration of treatments, providing patients who are most likely to benefit from extended therapies receive suitable care. This individualized strategy not only improves patient survival but also optimizes the allocation of public health resources, preventing unnecessary protracted treatments in low-risk patients.
In summary, protracted ATB therapy offers a substantial survival advantage for patients with hematologic malignancies who are at high risk. This study emphasizes the impact of applying accurate risk stratification models in determining decisions regarding treatment, thereby improving patient outcomes through tailored medical care. Future research should prioritize the validation of these findings in a more expansive, multicenter investigations and explore the fundamental mechanisms contributing to the improved outcomes found in high-risk individuals.
The research indicates that the administration of granulocyte-colony stimulating factor (G-CSF) markedly affects the absolute neutrophil count (ANC) in patients undergoing prolonged antibiotic treatment for bloodstream infections associated with hematologic malignancies (HM). The mean ANC was lower in the extended ATB group than in the 14-day ATB group; nevertheless, statistical analysis revealed no significant difference in ANC levels between the groups (p = 0.5492, t-test). This indicates that G-CSF administration influences ANC but does not obscure the established correlation between prolonged antibiotic treatment and clinical outcomes.
Clinicians' decisions about G-CSF administration are frequently affected by baseline ANC levels, with patients exhibiting low ANC becoming more likely to receive G-CSF treatment following the onset of febrile episodes. This underscores the adaptive role of supportive care in the management of HM, wherein G-CSF is used to alleviate neutropenia and reduce complications related to infections.
Prior researches (Clark et al., 2005; Estcourt et al., 2016) revealed that the administration of G-CSF shortens the duration of neutropenia; however, it does not have a significant impact on overall mortality rates. This finding corresponds with our study results, indicating that the difference in G-CSF administration between high-risk (45.8%) and low-risk (25.9%) patients did not result in significant variations in mortality. This points out that although G-CSF can promote neutrophil recovery, its impact on long-term outcomes may be restricte.
These observations emphasize the necessity of accounting for baseline ANC and clinical status in the assessment of G-CSF effects in forthcoming studies. The confounding effects of G-CSF administration, along with its differential impact on neutrophil recovery and patient outcomes, require precise stratification and standardized protocols in research. Furthermore, multicenter studies are required to evaluate the generalizability of these findings across various patient populations and supportive care strategies.
Multiple aspects of this study stand out as strengths, including the methodological rigor, the clinical relevance, and the robustness of the Ex-CSEPA model. First, the methodological rigor: the study applied a rigorous methodology by conducting a full multivariable regression analysis. This analysis accounted for multiple prognostic factors in order to evaluate the effectiveness of extended antibiotic therapy. This methodology ensured that the results were strong and obtained into consideration all possible factors that could affect the outcome. Second, the robustness of model: The Ex- CSEPA model demonstrated outstanding resilience by accurately categorizing patients into high-risk and lowrisk cohorts, resulting in notable disparities in survival rates. The accuracy of the model in estimating the probability of mortality within 30 days was confirmed using several analytical techniques, such as propensity score analysis and Weibull proportional hazard regression. Third, the clinical relevance: the results have immediately apparent and crucial consequences for the application of medical treatments. The study establishes a foundation for personalized treatment strategies that could improve patient outcomes and optimize the consumption of medical resources by identifying highrisk patients who benefit from extended ATB therapy. The model's capacity to guide clinical decision-making underlines the potential for usefulness in routine clinical settings. Forth, implications for clinical practice: The study highlights the capacity of extended ATB therapy to considerably decrease the risk of mortality in highrisk patients indicated by the Ex-CSEPA model. The personalized strategy of prolonged antibiotic therapy for individuals with a greater likelihood of suffering from negative consequences emphasizes the importance of tailored medical care. In order to enhance patient outcomes and optimize resource allocation, clinicians may recognize and focus on high-risk patients with a more intensive treatment strategy. In clinical practice, the findings underscore the urgent necessity of precise risk stratification tools, such as the Ex-CSEPA model. These models empower healthcare practitioners to make well-informed decisions regarding the duration of treatments, ensuring that patients who are most likely to benefit from extended therapies receive appropriate care. This individualized strategy not only improves patient survival but also optimizes the allocation of public health resources, preventing unnecessary prolonged treatments in low-risk patients.
Despiteitsstrengths,thisstudyhasseveral limitations. First, the generalizability of the findingsis restricted by the cohort size of 82 patients, which requires the performance of larger studies to verify these findings. Second, the retrospective nature of the study may be susceptible to data accuracy issues and unmeasured confounding factors, while the singlecenter design may introduce bias related to specific clinical practices and patient populations. Third, while the study highlights the applicability of the Ex-CSEPA model in predicting mortality among patients with CRAB infections, its generalizability to other bacterial infections remains uncertain due to the prevalence of CRAB infections in the cohort. Additionally, the use of nephrotoxic antibiotics, such as colistin, contributed to a higher incidence of acute kidney injury (AKI) in the extended antibiotic therapy group, potentially influencing the observed outcomes.
The potential sources of bias in score calculation, the limited diversity of the cohort, and the unmeasured confounding factors, including differences in supportive care practices or emerging pathogen resistance patterns, may also affect the predictive accuracy of the model. Moreover, pre-existing kidney disease was accounted for in the multivariable regression analysis and through the inclusion of the Charlson comorbidity score, which considers chronic kidney disease as a factor. However, these elements underline the need for cautious interpretation of the findings.
Future research should prioritize the generalizability and robustness of the Ex-CSEPA model by conducting larger, multicenter investigations across diverse patient populations to validate and refine its predictive capabilities. Moreover, exploring the underlying mechanisms behind the observed benefits of extended antibiotic therapy in high-risk patients could enhance our understanding of its clinical utility. By addressing these limitations, the Ex-CSEPA model could be further strengthened, enabling more precise risk stratification and effective personalized treatment strategies in clinical practice.
CONCLUSION
The Ex-CSEPA model exhibits significant predictive capability for assessing 30-day mortality risk in patients with hematologic malignancies receiving a typical 14-day antibiotic regimen for AB bacteremia. The approach efficiently stratifies patients into high-risk and low-risk groups, revealing considerable disparities in prognostic variables associated with mortality. The 30-day survival analysis indicates a significant difference, as the high-risk group demonstrates a poorer median survival than low-risk individuals, so confirming the model's effectiveness in predicting patient outcomes for this population.
Extended antibiotic therapy (ATB) provides a substantial survival benefit for those identified as highrisk, emphasizing the value of precise risk stratification models in guiding individualized treatment strategies. This study underscores the Ex-CSEPA model's potential in clinical decision-making, resulting to improved patient outcomes through personalized medical care. Future research should prioritize validating these findings across larger, multicenter cohorts that encompass diverse patient populations. Moreover, integrating the Ex-CSEPA model with other clinical risk scores or predictive tools for hematologic malignancies might provide new insights, potentially broadening its applicability and clinical impact. Exploring the fundamental molecular or immunological underpinnings of prolonged antibiotic therapy may reveal innovative therapeutic approaches. Additionally, utilizing emerging technologies like machine learning could improve the model's predictive accuracy, ensuring its adaptability to changing pathogen resistance patterns and differences in clinical practice. Ultimately, the development of precision medicine in the management of infectious diseases among patients with hematologic conditions will be facilitated by these advancements.
ACKNOWLEDGEMENTS
For his invaluable assistance and unwavering encouragement throughout this research, we would like to extend my heartfelt gratitude to the Academic Service and Research Institute (UNISEARCH), Payap University.
REFERENCES
- Allegra A, Tonacci A, Musolino C, Pioggia G, Gangemi S. SecondaryImmunodeficiencyin Hematological Malignancies: Focus on Multiple Myeloma and Chronic Lymphocytic Leukemia. Front Immunol. 2021; 12:738915.
- Azam F, Latif MF, Farooq A, Tirmazy SH, AlShahrani S, Bashir S, et al. Performance Status Assessment by Using ECOG (Eastern Cooperative Oncology Group) Score for Cancer Patients by Oncology Healthcare Professionals. Case Rep Oncol. 2019;12(3):728-36.
- Baker TM, Satlin MJ. The growing threat of multidrugresistant Gram-negative infectionsinpatientswith hematologic malignancies. Leuk Lymphoma. 2016;57(10):2245-58.
- Chaiwarit R, Boonchuay M, Supparatpinyo K, Srisanthana T. Pandrug-Resistant Acinetobacter baumannii at Maharaj Nakorn Chiang Mai Hospital. J Infect Dis Antimicrob Agents. 2005;22(108).
- Chandra S, Kalashnikova T, Wright NAM, Dávila Saldaña BJ. Primary Immunodeficiencies and Hematologic Malignancies: A Diagnostic Approach. Front Immunol. 2022;13.
- Clark OA, Lyman GH, Castro AA, Clark LG, Djulbegovic B. Colony-stimulating factors for chemotherapy-induced febrile neutropenia: a metaanalysis of randomized controlled trials. J Clin Oncol. 2005;23(18):4198-214.
- Contejean A, Lemiale V, Resche-Rigon M, Mokart D, Pène F, Kouatchet A, et al. Increased mortality in hematological malignancy patients with acute respiratory failure from undetermined etiology: a Groupe de Recherche en Réanimation Respiratoire en Onco-Hématologique (Grrr-OH) study. Ann Intensive Care. 2016;6(1):102.
- De Waele JJ, Martin-Loeches I. Optimal duration of antibiotic treatment in Gram-negative infections. Curr Opin Infect Dis. 2018;31(6):606-11.
- Estcourt LJ, Stanworth SJ, Hopewell S, Doree C, Trivella M, Massey E. Granulocyte transfusions for treating infections in people with neutropenia or neutrophil dysfunction. Cochrane Database Syst Rev. 2016;4(4):Cd005339.
- Garcia-Vidal C, Cardozo-Espinola C, Puerta-Alcalde P, Marco F, Tellez A, Agüero D, et al. Risk factors for mortality in patients with acute leukemia and bloodstream infections in the era of multiresistance. PLoS One. 2018;13(6):e0199531.
- Im Y, Kang D, Ko R-E, Lee YJ, Lim SY, Park S, et al. Time-to-antibiotics and clinical outcomes in patients with sepsis and septic shock: a prospective nationwide multicenter cohort study. Crit Care. 2022;26(1):19.
- Katip W, Uitrakul S, Oberdorfer P. Short-Course Versus Long-Course Colistin for Treatment of Carbapenem- Resistant A. baumannii in Cancer Patient. Antibiotics (Basel). 2021;10(5).
- Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med. 1985;13(10):818-29.
- Lopes JA, Jorge S. The RIFLE and AKIN classifications for acute kidney injury: a critical and comprehensive review. Clin Kidney J. 2013;6(1):8-14.
- Nørgaard M, Larsson H, Pedersen G, Schønheyder HC, Sørensen HT. Risk of bacteraemia and mortality in patients with haematological malignancies. Clin Microbiol Infect. 2006;12(3):217-23.
- Park K-H, Jung YJ, Lee HJ, Kim HJ, Maeng CH, Baek SK, et al. Impact of multidrug resistance on outcomes in hematologic cancer patients with bacterial bloodstream infections. Sci Rep. 2024;14(1):15622.
- Plackoska V, Shaban D, Nijnik A. Hematologic dysfunction in cancer: Mechanisms, effects on antitumor immunity, and roles in disease progression. Front Immunol. 2022;13.
- Rhee J-Y, Kwon KT, Ki HK, Shin SY, Jung DS, Chung D-R, et al. Scoring System for Prediction of Mortality in Patients with Intensive Care Unit-Acquired Sepsis: A Comparison of the Pitt Bacteremia Score and the Acute Physiology And Chronic Health Evaluation II Scoring Systems. Shock. 2009;31(2):146-50.
- Sørensen JB, Klee M, Palshof T, Hansen HH. Performance status assessment in cancer patients. An inter-observer variability study. Br J Cancer. 1993;67(4):773-5.
- Sousa A, Pérez-Rodríguez MT, Suárez M, Val N, Martínez-Lamas L, Nodar A, et al. Shortversus longcourse therapy in gram-negative bacilli bloodstream infections. Eur J Clin Microbiol Infect Dis. 2019;38(5):851-7.
- Tansarli GS, Andreatos N, Pliakos EE, Mylonakis E. A Systematic Review and Meta-analysis of Antibiotic Treatment Duration for Bacteremia Due to Enterobacteriaceae. Antimicrob Agents Chemother. 2019;63(5).
- Tong-Minh K, Welten I, Endeman H, Hagenaars T, Ramakers C, Gommers D, et al. Predicting mortality in adult patients with sepsis in the emergency department by using combinations of biomarkers and clinical scoring systems: a systematic review. BMC Emerg Med. 2021;21(1):70.
- Turkoglu M, Mirza E, Tunçcan Ö G, Erdem GU, Dizbay M, Yağcı M, et al. Acinetobacter baumannii infection in patients with hematologic malignancies in intensive care unit: risk factors and impact on mortality. J Crit Care. 2011;26(5):460-7.
- Vaquero-Herrero MP, Ragozzino S, Castaño-Romero F, Siller-Ruiz M, Sánchez González R, García-Sánchez JE, et al. The Pitt Bacteremia Score, Charlson Comorbidity Index and Chronic Disease Score are useful tools for the prediction of mortality in patients with Candida bloodstream infection. Mycoses. 2017;60(10):676-85.
- Weinstein MP, Lewis JS, 2nd. The Clinical and Laboratory Standards Institute Subcommittee on Antimicrobial Susceptibility Testing: Background, Organization, Functions, and Processes. J Clin Microbiol. 2020;58(3).
Edited by
-
Associate Editor:
Silvya Stuchi Maria-Engler
Source: Prepared by the authors.