Karolina Gaebe

Ms. Karolina Gaebe

University of Toronto
orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
2 January 2024 in eClinicalMedicine
Verified
Journal Article
Open access
KGKarolina Gaebe
Karolina Gaebe
AEAnders W. Erickson
Anders W. Erickson
ALAlyssa Y. Li
Alyssa Y. Li
AYAndrew N. Youssef
Andrew N. Youssef
BSBhagyashree Sharma
Bhagyashree Sharma
SDSunit Das
Sunit Das
Patients with small cell lung cancer (SCLC) are at high risk for brain metastases. Prophylactic cranial irradiation (PCI) is recommended in this population to reduce the incidence of brain metastases and prolong survival. We aimed to assesses the efficacy of PCI in this population in the era of routine brain imaging. To our knowledge, this is the first systematic review and meta-analysis to examine the use among patients who were radiographically confirmed not to have brain metastases after completion of first-line therapy. In this systematic review and meta-analysis, cohort studies and controlled trials reporting on the use of PCI for patients SCLC were identified in EMBASE, MEDLINE, CENTRAL, and grey literature sources. The literature search was conducted on November 12, 2023. Summary data were extracted. Random-effects meta-analyses pooled hazard ratios (HR) for the primary outcome of overall survival between PCI and no intervention groups. This study is registered with the Open Science Framework, DOI:10.17605/OSF.IO/BC359, and PROSPERO, CRD42021249466. Of 4318 identified records, 223 were eligible for inclusion. 109 reported on overall survival in formats amenable to meta-analysis; PCI was associated with longer survival in all patients with SCLC (HR 0.59; 95% CI, 0.55–0.63; p < 0.001; n = 56,770 patients), patients with limited stage disease (HR 0.60; 95% CI, 0.55–0.65; p < 0.001; n = 78 studies; n = 27,137 patients), and patients with extensive stage disease (HR 0.59; 95% CI, 0.51–0.70; p < 0.001; n = 28 studies; n = 26,467 patients). Between-study heterogeneity was significant when pooled amongst all studies (I2 = 73.6%; 95% CI 68.4%–77.9%). Subgroup analysis did not reveal sources of heterogeneity. In a subgroup analysis on studies that used magnetic resonance imaging to exclude presence of brain metastases at restaging among all patients, overall survival did not differ significantly between patients who did or did not receive PCI (HR 0.74; 95% CI, 0.52–1.05; p = 0.08; n = 9 studies; n = 1384 patients). Our findings suggested that administration of PCI is associated with a survival benefit, but not when considering studies that radiographically confirmed absence of brain metastases, suggesting that the survival benefit conferred by PCI might be therapeutic rather than prophylactic. No funding.

3 October 2024 in eClinicalMedicine
Verified
Journal Article
Open access
KGKarolina Gaebe
Karolina Gaebe
AEAnders W. Erickson
Anders W. Erickson
SCSarah Chen
Sarah Chen
IMInes B. Menjak
Ines B. Menjak
BLBenjamin H Lok
Benjamin H Lok
SDSunit Das
Sunit Das
Patients with small cell lung cancer (SCLC) have historically been characterised by poor overall survival (OS) and high risk for brain metastasis (BM), but large-scale real-world evidence on clinical presentation and treatment in this population is lacking. Our aim was to describe the clinical characteristics and outcomes of patients with SCLC and BM in Ontario, Canada. This population-based, retrospective cohort study included all patients in Ontario, Canada, who were diagnosed with SCLC between April 1, 2010, and March 31, 2018. Data were analysed between June 2, 2022, and December 20, 2023. Patients with second cancer diagnosis were excluded. Patients were identified and data retrieved from the Institute for Clinical Evaluative Sciences (ICES) databases. Kaplan–Meier and multivariable Cox regression analyses were performed to compare OS between patient cohorts stratified by disease stage, BM diagnosis, and intracranial treatment modality. Propensity score-matching based on age, disease stage, time to BM, and receipt of chemotherapy was performed to compare OS between intracranial treatment modalities. 8705 patients were included (male: 4433, female: 4272). Median age at diagnosis was 68 years (interquartile range, IQR, 61–75). Median OS of all patients was 7.46 months (95% confidence interval, CI, 7.23–7.69). 32% (n = 2686) of patients developed BM (synchronous, 43.7%; asynchronous, 56.3%) with median OS of 9.76 months (95% CI, 9.36–10.22). 102 (4%), 1654 (62%), and 930 (35%) patients received stereotactic radiosurgery (SRS), whole brain radiation therapy (WBRT), or no treatment, respectively, for their BM in the first-line setting or after prophylactic cranial irradiation (PCI). In propensity score-matched analyses, OS from time of BM diagnosis was non-inferior between SRS- and WBRT-treated cohorts among patients who did not receive PCI (hazard ratio, HR, 0.68, 95% CI, 0.44–1.06, p = 0.091, n = 86) and in favour of SRS for those who received PCI prior to BM development (HR, 0.47, 95% CI, 0.31–0.72, p = 0.0042, n = 112). OS for patients with SCLC remains poor, and many patients present with BM. With careful selection, patients with SCLC and BM may benefit from SRS treatment. Future research should incorporate information on burden of intracranial disease and novel immunotherapies. None.

7 March 2022 in Frontiers in Oncology
Verified
Journal Article
Open access
ALAlyssa Y. Li
Alyssa Y. Li
KGKarolina Gaebe
Karolina Gaebe
KJKatarzyna Jerzak
Katarzyna Jerzak
PCParneet K. Cheema
Parneet K. Cheema
ASArjun Sahgal
Arjun Sahgal
SDSunit Das
Sunit Das
Intracranial metastatic disease (IMD) is a prevalent complication of cancer that significantly limits patient survival and quality of life. Over the past half-century, our understanding of the epidemiology and pathogenesis of IMD has improved and enabled the development of surveillance and treatment algorithms based on prognostic factors and tumor biomolecular characteristics. In addition to advances in surgical resection and radiation therapy, the treatment of IMD has evolved to include monoclonal antibodies and small molecule antagonists of tumor-promoting proteins or endogenous immune checkpoint inhibitors. Moreover, improvements in the sensitivity and specificity of imaging as well as the development of new serological assays to detect brain metastases promise to revolutionize IMD diagnosis. In this review, we will explore current treatment principles in patients with IMD, including the emerging role of targeted and immunotherapy in select primary cancers, and discuss potential areas for further investigation.

1 August 2025 in PLOS ONE
Verified
Journal Article
Open access
PDF file
KGKarolina Gaebe
Karolina Gaebe
BWBenjamin van der Woerd
Benjamin van der Woerd
Background: Large language models (LLMs) have demonstrated capabilities in natural language processing and critical reasoning. Studies investigating their potential use as healthcare diagnostic tools have largely relied on proprietary models like ChatGPT and have not explored the application of advanced prompt engineering techniques. This study aims to evaluate the diagnostic accuracy of three open-source LLMs and the role of prompt engineering using clinical scenarios. Methods: We analyzed the performance of three open-source LLMs—llama-3.1-70b-versatile, llama-3.1-8b-instant, and mixtral-8x7b-32768—using advanced prompt engineering when answering Medscape Clinical Challenge questions. Responses were recorded and evaluated for correctness, accuracy, precision, specificity, and sensitivity. A sensitivity analysis was conducted presenting the three LLMs with basic prompting challenge questions and excluding cases with visual assets. Results were compared with previously published performance data on GPT-3.5. Results: Llama-3.1-70b-versatile, llama-3.1-8b-instant, and mixtral-8x7b-32768 achieved correct responses in 79%, 65%, and 62% of cases, respectively, outperforming GPT-3.5 (74%). Diagnostic accuracy, precision, sensitivity, and specificity responses all outperformed those previously reported for GPT-3.5. Results generated using advanced prompting strategies were superior to those based on basic prompting. Sensitivity analysis revealed similar trends when cases with visual assets were excluded. Discussion: Using advanced prompting techniques, LLMs can generate clinically accurate responses. The study highlights the limitations of proprietary models like ChatGPT, particularly in terms of accessibility and reproducibility due to version deprecation. Future research should employ prompt engineering techniques and prioritize the use of open-source models to ensure research replicability.

12 May 2025 in The Laryngoscope
Verified
Journal Article
Open access
KGKarolina Gaebe
Karolina Gaebe
ANAmy Ng
Amy Ng
JSJennifer L. Spiegel
Jennifer L. Spiegel
TLTrung N. Le
Trung N. Le
VLVincent Y.‐W. Lin
Vincent Y.‐W. Lin
JCJoseph M. Chen
Joseph M. Chen
Objective: To assess the financial benefits and carbon emission savings associated with performing virtual‐synchronous remote programming and troubleshooting following cochlear implant (CI) placement for patients who live in remote regions.Methods: CI users living ≥ 100 km from the implant center with ≥ 1 month CI experience were included. Participants were either assigned to home visits using a portable laptop (Remote Laptop; n = 20) or attended visits through the nearest Remote Hosted Site (n = 34). Direct costs in Canadian dollars from a Ministry of Health perspective and carbon emissions were compared between remote and in‐person appointments at the implant center.Results: Median distances between participants' homes to the implant center were 574 km (Remote Laptop) and 999 km (Remote Hosted Site). Average costs for both groups of remote care were lower compared with in‐person appointments at the implant center (Remote Laptop: $180.70 vs. $431.61, p = 0.002; Remote Hosted Site: $155.34 vs. $1110.89, p < 0.001). Total cost reductions were $5018.23 and $32,488.76 with cost‐to‐benefit ratios of 2.39 and 7.15 for Remote Laptop and Remote Hosted Site groups, respectively. Net carbon emission benefits were 2200.00 kg and 5691.50 kg with net carbon benefit ratios of 512.63 and 5.25 for Remote Laptop and Remote Hosted Site, respectively.Conclusion: Virtual synchronous remote CI programming and troubleshooting generates cost savings in post‐surgical care. Additionally, it achieves remarkable carbon emissions savings. More widespread adoption is expected, especially for those who reside far from implant centers.

2 January 2025 in The Laryngoscope
Verified
Journal Article
Open access
NVNoémie Villemure‐Poliquin
Noémie Villemure‐Poliquin
RFRui Fu
Rui Fu
KGKarolina Gaebe
Karolina Gaebe
JKJin Kwon
Jin Kwon
MCMarc Cohen
Marc Cohen
AEAntoine Eskander
Antoine Eskander
Objectives: To evaluate the impact of delayed postoperative radiotherapy (PORT) on overall survival (OS) in patients with head and neck cancers (HNC). Data Sources: A systematic review and meta‐analysis were conducted by searching MEDLINE, Embase, CENTRAL, Web of Science, and CINAHL databases. Review Methods: Studies assessing the impact of delayed PORT in adult HNC patients were included. A total of 11,171 titles and abstracts were screened, with 52 studies meeting the inclusion criteria. Data were extracted, and a pooled random‐effects analysis was performed. The primary outcome was overall survival (OS), comparing patients receiving timely PORT (within 42 days) to those with delays. Results: Of the included studies, 31 were conducted in the United States, with 16 using the National Cancer Database (NCDB). Patients who did not receive PORT within 42 days had a 4% increase in mortality (adjusted Hazard Ratio [aHR]: 1.04 [1.03–1.06]; I 2 = 78%; N = 254,189; 16 studies). Excluding time‐overlapping NCDB‐based studies, the OS benefit for timely treatment persisted (aHR: 1.10 [1.01–1.20]; I 2 = 39%; N = 52,003; 5 studies). Conclusions: Initiating PORT within 42 days is significantly associated with decreased mortality in HNC patients, reinforcing CoC recommendations. However, more research is needed to understand the relationship between different time cutoffs and outcomes, and to identify factors contributing to PORT delays. Future studies should explore the impact of treatment delays on patient‐centered outcomes, such as Quality of Life (QoL). Level of Evidence: NA Laryngoscope , 135:1563–1570, 2025