Flavia Dei Zotti

Flavia Dei Zotti

Auto-generated profile, based on publicly available information

Flavia Dei Zotti

Department of Pathology and Cell Biology, Columbia University Irving Medical Center, New York, NY

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

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

Representative Publications

View All Publications
10 October 2024 in Blood
Verified
Journal Article
Closed access
FZFlavia Dei Zotti
Flavia Dei Zotti
AQAnnie Qiu
Annie Qiu
VDVivette D. D’aGati
Vivette D. D’aGati
SJShwatina Jagnarine
Shwatina Jagnarine
EKEmmalene Kyritsis
Emmalene Kyritsis
KHKrystalyn E. Hudson
Krystalyn E. Hudson
Immune checkpoint inhibitors (ICPis) have revolutionized cancer immunotherapy but also can induce autoimmune hemolytic anemia (AIHA), a severe disease with high mortality. However, the cellular and molecular mechanism(s) of AIHA secondary to ICPi therapy (ICPi-AIHA) are unclear, other than being initiated through decreased checkpoint inhibition. Herein, we report ICPi-AIHA in a novel mouse model that shows similar characteristics of known human ICPi-AIHA (eg, autoantibodies, hemolysis, and increased mortality). During ICPi-AIHA, there is the simultaneous reduction of 2 regulatory T-cell populations (FoxP3+ and Tr1 [type 1 regulatory cells]) and an increase in inflammatory T helper cell 17 (TH17). Moreover, a novel CD39+CD73–FoxP3–CD25– CD4+ T-cell subset (ie, CD39 single positive [CD39SP]) emerges, and early increases in CD39SP predict AIHA development; CD39 is an ectonuclease that breaks down adenosine triphosphate (ATP). Additionally, we found that boosting ATPase activity by injecting recombinant apyrase mitigates AIHA development and significant CD39SP reductions, both suggesting a functional role for CD39 and demonstrating a novel therapeutic approach. Importantly, CD39SP are detectable in multiple mouse models developing AIHA and in patients with AIHA, demonstrating applicability to idiopathic and secondary AIHA. Highlighting broader autoimmunity relevance, ICPi-treated NZB mice experienced accelerated onset and severity of lupus, including AIHA. Moreover, ICPi treatment of healthy B6 animals led to detectable CD39SP and development of autoantibodies against multiple autoantigens including those on red blood cells and platelets. Together, our findings provide further insight into the cellular and molecular mechanisms of ICPi-AIHA, leading to novel diagnostic and therapeutic approaches with translational potential for use in humans being treated with ICPi.

30 October 2026 in British Journal of Haematology
Verified
Journal Article
Closed access
PDF file
FZFlavia Dei Zotti
Flavia Dei Zotti
SJShwatina Jagnarine
Shwatina Jagnarine
EKEmmalene Kyritsis
Emmalene Kyritsis
WZWei Zhu
Wei Zhu
KHKrystalyn E. Hudson
Krystalyn E. Hudson

1 October 2025 in bioRxiv
Verified
Preprint
Open access
AGAndrea Gazzo
Andrea Gazzo
SBSilvia Berardelli
Silvia Berardelli
MBMatteo Biancospino
Matteo Biancospino
LCLorenzo Cuollo
Lorenzo Cuollo
FZFlavia Dei Zotti
Flavia Dei Zotti
Paolo MagniPaolo Magni
Paolo Magni
Background: Generative AI is increasingly used to extract structured information across domains, but its reliability in academic and clinical research, where precision and accuracy are essential, remains largely unexplored. This study evaluates the ability of Large Language Models (LLMs)-based algorithms to generate accurate, literature-based summaries of human genomic variants, with a focus on real-world usability. Results: We benchmarked five open-access generative AI platforms—ChatGPT, MistralAI, VarChat, Perplexity, and ScholarAI—across 40 curated variants equally divided between somatic and germline settings. For each variant, summary reports were generated and blindly evaluated by domain experts using five defined metrics. VarChat emerged as the top-ranked tool, showing the highest summarization accuracy, citation relevance, and robustness against hallucinations. Gpt-4o consistently ranked second, showing particularly stable robustness in conditions where the literature was scarce. Perplexity and ScholarAI, despite being literature-focused, ranked lowest across most metrics. Tool performance was strongly influenced by the availability of peer-reviewed literature, confirming that current generative models remain sensitive to data scarcity. Conclusions: Our findings highlight the heterogeneity of current generative AI tools in genomic variant interpretation workflows. While some platforms already provide useful outputs, reliable integration into basic and clinical research requires expert validation and domain-related fine-tuning. This work provides for the first time a curated benchmark for assessing LLM-generated content in variant genomics and underscores the need for caution when using these tools to support variant interpretation.