Page 56 - ITPS-8-3
P. 56

INNOSC Theranostics and
            Pharmacological Sciences                                           Precision medicine and beyond in oncology



            model that uses genomic profile and xenograft model data   for AI as a tool to be more patient-specific with improved
            from pre-clinical and biomedical studies to predict clinical   clinical outcomes.  We also believe that AI will eventually
            outcomes of drug combinations for cancer patients with   be  incorporated into medical practices.  However, we
            comorbid conditions such as type II diabetes.  Although   would  like to  acknowledge significant challenges,  such
                                                146
            this particular tool is in its early stages of development,   as logistical challenges to digital pathology, data quality
            innovations such as PINNED, MADRIGAL, and AlphaFold   concerns, risk of bias, ethical implications, and potential
            are indicative of AI’s potential in pharmaceutical research.  compromise to patient trust. The current advancement of
                                                               AI in medicine accompanies a worrying lack of legislative
            5.3. AI in clinical practice                       framework, making it susceptible to breaches of patient
            Some studies have explored the idea of using AI to inform   rights. Therefore, rash adoption of powerful technology
            clinical decision-making. A  recent study from China   is like building a new story on a foundation of quicksand,
            assessed the impact of an AI-based clinical decision   which will lead to an ultimate collapse. We are optimistic
            support system on the treatment of breast cancer patients.    that with careful planning and thoughtful implementation,
                                                         147
            A  group  of  physicians  was  asked  to  provide  treatment   AI can be incorporated in a way that truly benefits patients
            recommendations for an average of 198  patients before   while minimizing unintentional harm.
            and after viewing individualized AI-generated treatment
            plans. Researchers found that adherence to National   Acknowledgments
            Comprehensive  Cancer  Network guidelines increased   The authors acknowledge the Biorender.com website used
            slightly (0.5%;  p=0.003) after the implementation of AI   to create Figures 1 and 2.
            support for the treatment of patients with stages I–III
            breast cancer. 147                                 Funding
              Clinicians can also use AI-based knowledge graphing   None.
            and summary tools to stay updated on the most recent
            developments  in  cancer  treatment.  Chandak  et al.    Conflict of interest
                                                         148
            presented an AI-based multimodal knowledge  graph,   The authors declare that they have no competing interests.
            PrimeKG, which synthesized data from 20 primary
            databases to map relationships between the proteins,   Author contributions
            genes, phenotypes, and risk factors associated with over   Conceptualization: Linwei Li
            17,000 diseases, including cancers.  PrimeKG’s integrative   Visualization: Anjali Binoy, Sriya Gullapalli
                                       148
            network also describes indications, contraindications, and   Writing – original draft: Linwei Li, Annu Karithara, Angel
            off-label uses of drugs used to treat these diseases. 148
                                                                  Phillip, Jennifer Escamilla, Anika Doppalapudi,
              Several large language model AI tools capable of    Christine Pham
            generating human-like text are in development for clinical   Writing – review & editing: Lois Baldado, Kaitlyn Ybanez,
            applications. However, this particular area of AI research is   Daniela Ramos, David Sta. Maria, Arjun Bellamkonda,
            still in its infancy and requires further investigation before   Amin Ibrahim, Hugo Zamarron, Daniela Gonzalez,
            large language model tools can be integrated into clinical   Shizue Mito, Linwei Li
            workflows. 145
                                                               Ethics approval and consent to participate
            6. Conclusion
                                                               Not applicable.
            This review discussed the potential of precision medicine
            in the field of oncology. The review has explored first-line   Consent for publication
            targeted therapies and immunotherapies that have been   Not applicable.
            well-established in the current standard care. Furthermore,
            the review discussed current targeted therapies in HCC,   Availability of data
            CRC, and PDAC and various trials of different targeted
            therapies and immunotherapies to achieve a more    Not applicable.
            efficacious regimen. Finally, this review has examined   References
            the emerging avenues in the field of precision medicine
            in cancer diagnosis, drug development, and clinical   1.   König IR, Fuchs O, Hansen G, von Mutius E, Kopp MV. What
            practice. We chose to discuss the possibility of including   is precision medicine? Eur Respir J. 2017;50(4):1700391.
            AI in our review, as we have seen the immense potential      doi: 10.1183/13993003.00391-2017



            Volume 8 Issue 3 (2025)                         50                          doi: 10.36922/ITPS025140018
   51   52   53   54   55   56   57   58   59   60   61