Ciara Kelly
Research Interests
Make your next move.
Prepare your application
Review drafts before you send them.
Explore related searches
Articles (30)
Purpose: The CDK4/6 inhibitor palbociclib delays progression in patients with advanced dedifferentiated liposarcoma (DDLPS). In carcinoma models, CDK4/6 and PD-1 inhibition induce intratumoral inflammation and synergistic activity. Comprehensive assessment of tumor and host dynamics is needed to understand response and resistance to this combination. Patients and Methods: We performed a phase 2 study of palbociclib and the PD-1 inhibitor retifanlimab in advanced DDLPS. Palbociclib was administered 2 weeks prior to retifanlimab. Tumor biopsies were analyzed by single-cell RNA sequencing (scRNA-seq), and peripheral blood by high-parameter flow cytometry. Analyses focused on cell cycle, senescence, and immune-related changes. Results: Twelve patients were treated before the study was halted because of a 42% immune-related toxicity rate. The overall response rate was 8.3%, and the disease control rate was 75%. Median progression-free and overall survival rates were 7.1 and 26.8 months, respectively. Nine patients had at least one tumor biopsy analyzed; three had paired. Ten patients had paired blood samples analyzed. Following treatment, tumor cells demonstrated decreased cycling and increased cell-cycle checkpoint activity. Transcriptional scores for senescence increased in cancer cells, as did the proportion of intratumoral T and B cells. A cluster of cells emerged with upregulated cell-cycle genes and downregulated HLA class I, suggesting innate or acquired resistance to treatment. In blood, a subset of CD4 + T cells decreased, whereas expression of LAG-3, ICOS, and CD38 increased in select subsets. Conclusions: A palbociclib lead-in prior to retifanlimab had a high rate of immune-related toxicities. Correlative analyses identified changes in tumor and immune cells attributable to treatment. A study of concurrent dosing of the combination is ongoing. Significance: In this phase 2 study of palbociclib and retifanlimab, treatment led to immune-related adverse events, halting enrollment. scRNA-seq and flow cytometry revealed cell-cycle suppression, senescence, and T-cell changes. These findings highlight the rationale for this combination and the need for safer dosing strategies in larger cohorts.
Year:
2026
Changes in tumor and immune cell populations and phenotypes with anti–PD-1. A, CDK4 and MDM2 expression in tumor cells according to best response. PD, P03, before and after retifanlimab initiation (IO); SD, three before IO and six after IO. B, Signature scores in tumor cells according to best response. Same samples as in A . C, Integrated clustering of cancer cells from patients with paired samples and sufficient tumor fraction. D and E, Intracluster signature score changes with initiation of retifanlimab, plotted ( D ) along the x -axis or ( E ) as dot size. F , Volcano plot of differential expression of genes related to immune response in cluster 8 vs. all other clusters from P12.
Year:
2026
Changes in proportions of T-cell populations with anti–PD-1 via high-parameter flow cytometry. A, Change in abundance of T-cell subtypes with initiation of retifanlimab in patients with paired biopsies and detectable tumor-infiltrating T cells. B, Boxplot depicting percent change post- vs. pre-retifanlimab in the abundance of 63 T cell types. Filled boxes with bold labels and asterisks indicate T cell subsets with significant changes ( P C and D, Percent change in four key subsets of ( C ) CD4 + and ( D ) CD8 + T cells. OR, overall response.
Year:
2026
Clinical trial design and efficacy. A, Study schema depicting treatment and biopsy collection time points. B, Flow diagram of treatment and follow-up for patients in the safety lead-in and trial expansion cohorts. C, Waterfall plot showing best percent tumor volume change from baseline per RECIST 1.1. One patient who was not evaluable for response by RECIST 1.1 is excluded. irAE(+), solid circles indicate irAEs; SC+, single-cell data from tumor biopsies, in which filling of the left and right halves indicates analyzability at the pre- and post-retifanlimab time points, respectively. D, Spider plot showing percent change in sum of target lesion diameters over time in patients who were evaluable for response by RECIST v1.1. E, Swimmer plot showing patient PFS. Dashed line indicates median PFS. NE, not evaluable; OR, overall response.
Year:
2026
Overview of scRNA-seq data. A, Cell clustering by UMAP. B, Expression of key markers within major cell populations, pooled among all specimens. C, Cell type proportion in each patient. Patients with paired biopsies were merged. D, Comparison of cell type proportions before vs. after retifanlimab initiation (IO) in patients with paired biopsies. UMAP, Uniform Manifold Approximation and Projection.
Year:
2026
Supplementary Figure 1. Traditional risk stratification in primary gastric and small bowel GIST. Primary gastric (A-C) and small bowel (D-F) GISTs localized at diagnosis without prior neoadjuvant treatment are risk stratified into low, moderate, and high risk groups based on site, tumor size and mitotic rate using the Miettinen (also includes very low risk) (A, D), Fletcher (B, E) and Joensuu (C, F) risk stratification schemes (Miettinen 2006, Fletcher 2002, Joensuu 2008). Recurrence-free survival for patients in each of these risk groups is represented by Kaplan-Meier curves (global log-rank P) for all primary localized cases.
Year:
2025
Supplementary Table S1. Metastasis Frequencies by Traditional vs. Genomic Risk Stratification Supplementary Table S2. List of gene level and chromosomal arm level genomic variables used in OncoCast. Supplementary Table S3-S4. Frequencies of genomic alterations by OncoCast risk groups in primary gastric and small bowel GISTs. Supplementary Table S5. Number and frequencies of specific KIT/PDGFRA mutations in primary localized gastric and small bowel GISTs.
Year:
2025
Supplementary Figure S4. RFS in primary localized gastric (A) and small bowel (B) GISTs by types of primary KIT and PDGFRA mutations [del: deletion; ex: exon; INS: insertion; PM: point mutation].
Year:
2025
Purpose: This study investigated factors that affected ctDNA detection and ctDNA as a molecular biomarker in a phase II trial of imatinib and binimetinib in newly diagnosed advanced gastrointestinal stromal tumor, including patients exposed to imatinib within 4 weeks of trial enrollment. Experimental Design: Plasma and tumor tissue samples were collected at baseline, during treatment, and upon progression. DNAs extracted from plasma and tumor tissue were analyzed using genomic assays, Memorial Sloan Kettering Analysis of Circulating cfDNA to Evaluate Somatic Status and Memorial Sloan Kettering Integrated Mutation Profiling of Actionable Cancer Targets, respectively. Sequenced ctDNA detection of the primary oncogenic driver was determined and correlated with clinical characteristics. Results: Patients ( n = 31) included in this analysis had KIT mutations ( n = 29, 94%) and metastatic disease ( n = 24, 77%) and achieved the best response of partial response ( n = 22, 71%), stable disease ( n = 8, 26%), or progressive disease ( n = 1, 3%). Sixteen patients (52%) were exposed to imatinib at baseline. The ctDNA detection rate of the primary oncogenic driver at baseline was 39% ( n = 12/31) and was significantly more likely in patients who were treatment-naïve ( n = 15) or had ≤4.2 weeks of treatment ( n = 8) than in those otherwise (48% vs. 13%; P value = 0.004). Baseline ctDNA detection did not correlate with tumor burden or stage. The ctDNA serial analysis of the primary oncogenic driver paralleled and sometimes preceded radiographic response. ctDNA detected resistance mutations in KIT ( n = 4). Active treatment influenced the detection of secondary KIT alterations in one patient. Conclusions: Active therapy at the time of ctDNA collection negatively affected the ability to detect primary and secondary KIT alterations in sequenced ctDNA from patients with advanced gastrointestinal stromal tumor. ctDNA responses may precede radiographic responses and merit further investigation.
Year:
2025
PURPOSE We hypothesized that ACR-368 (prexasertib) would be active in desmoplastic small round cell tumor (DSRCT) because of favorable responses in preclinical models. METHODS Preclinical work identified ACR-368 activity in DSRCT, and a phase I/II trial of ACR-368 and irinotecan in patients 12 months and older with relapsed/refractory DSRCT was conducted. The primary objectives were determination of recommended phase II dose (RP2D) and best overall response rate (ORR) at the RP2D in DSRCT, with ≥3 of 16 responses considered promising. RESULTS Preclinical data confirmed ACR-368 as potentially therapeutic in DSRCT, and 19 patients were enrolled in a subsequent clinical trial. Treatment was well tolerated, and cytopenias were managed using growth factors. Fifteen of 19 patients, including five of six achieving PR, had previously received irinotecan. The estimated ORR at the RP2D was 23% (lower boundary one-sided 90% CI, 9%), exceeding the unpromising rate of 5%. In addition, three patients with DSRCT had a PR at doses other than the RP2D, bringing the ORR for all doses (n = 19) to 32% (90% CI, 15% to 53%). The median overall survival was 19 months (95% CI, 13 to 36). CONCLUSION The RP2D of ACR-368 with irinotecan by age group is ACR-368 105 or 150 mg/m 2 once on day 1 (>21 years or ≤21 years, respectively) and irinotecan 15 mg/m 2 once daily for 5 days in 21-day cycles for both groups. The study met its primary objective to consider ACR-368 and irinotecan promising in DSRCT and, to our knowledge, is the first incorporating a targeted therapy to achieve this magnitude of response.
Year:
2025
Year:
2025
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2022
Collaborators (18)
Julia Glade Bender
Vice Chair for Clinical Research
Memorial Sloan Kettering Cancer Center
Mark A. Dickson
-
Samuel Singer
Vincent Astor Chair of Clinical Research
Memorial Sloan Kettering Cancer Center
Gerald Behr
Memorial Sloan Kettering Cancer Center
Raghu Kataru
Senior Research Scientist
Memorial Sloan Kettering Cancer Center
Nguan Soon Tan
Professor
Nanyang Technological University (NTU)
Rodrigo Gularte Mérida
Senior Research Scientist
Memorial Sloan Kettering Cancer Center
Sam S. Yoon
-
Josephine Kam Tai Dermawan
Associate Staff
Cleveland Clinic
Andrew Kung
Chair
Memorial Sloan Kettering Cancer Center
Benjamin A Nacev
University of Pittsburgh
Alex Kentsis
Memorial Sloan Kettering Cancer Center
Robert G. Maki
Professor, Member
-
Jason Chan
Cedars-Sinai Medical Center
Marta Kovatcheva
Group Leader (Assistant Professor Equivalent)
IFOM – The AIRC Institute of Molecular Oncology
Andrew Koff
Sloan Kettering Institute
Edmund Bartlett
Memorial Sloan Kettering Cancer Center
Violeta Serra
Group Leader
Vall d'Hebron Institute of Oncology
Make your next move.
Prepare your application
Review drafts before you send them.