Advances in Gastrointestinal Cancer Treatment: Insights from ASCO 2024

Non-Sponsored Content

Gastrointestinal (GI) cancers remain a global health challenge, with ongoing research continually seeking to improve diagnostic and therapeutic strategies. The American Society of Clinical Oncology (ASCO) 2024 conference showcased several groundbreaking studies that leverage technological advancements; this piece examines three critical studies from the conference, emphasising their contributions to the advancement of the field.

Efficacy and Safety of Neoadjuvant Immunotherapy in dMMR/MSI-H GI Cancers

Real-world data obtained from the DREAM-GI national database were used to evaluate the outcomes, safety, and response patterns of patients with deficient mismatch repair/microsatellite instability-high (dMMR/MSI-H) GI cancers undergoing neoadjuvant immunotherapy (NIT).¹ This initial report includes 50 patients with a median age of 67 years. The cohort consisted mostly of colorectal cancer (62%), with gastroesophageal (14%) and pancreaticobiliary cancers (24%) also represented.¹

Pembrolizumab monotherapy was the primary NIT regimen (80%), with some patients receiving ipilimumab plus nivolumab (16%) or nivolumab plus chemotherapy (4%).¹ The median duration of NIT was 6 months, with the time-to-best-response observed at 3 months.¹ Among 47 evaluable patients, the overall response rate was 75%, including 43% achieving a radiologic complete response (CR).¹ Stable and progressive diseases (PD) were observed in 15% and 10% of patients, respectively.¹ All patients with CR remained progression-free (n=24) after a 25.5-month median follow-up.¹ Only 10% (n=5) of patients underwent surgery post-NIT, of which 80% (n=4) achieved pathologic CR.¹ Twelve patients passed due to disease progression, immunotherapy-related pneumonitis, and comorbidities.¹ This study showed a remarkable and durable response in patients with dMMR/MSI-H GI cancers receiving NIT despite infrequent PD, though further exploration is warranted.

Achieving the Best Outcomes under COVID-19 Endemicity

Inactivated and mRNA COVID-19 vaccines in patients with autoimmune rheumatic diseases – A city of two tales

Late but sinister post-COVID-19 conditions: new-onset rheumatic diseases

COVID-19 Infection and Subfertility: A Subtle and Debatable Complication

Previous Next

17 May 2023

Achieving the Best Outcomes under COVID-19 Endemicity

30 January 2023

Inactivated and mRNA COVID-19 vaccines in patients with autoimmune rheumatic diseases – A city of two tales

Disparities and Trends in Biomarker Testing in Metastatic Colorectal Cancer

Another study addressed the utilisation and disparities in biomarker testing for metastatic colorectal cancer (mCRC). Biomarker testing is crucial for targeted therapies, yet the guideline-recommended implementation of RAS, BRAF, and microsatellite instability/mismatch repair (MSI/MMR) testing may vary. The study analysed data from the Flatiron Health electronic health record database, encompassing 25,937 mCRC patients diagnosed post-January 2013 and evaluated testing rates for KRAS, NRAS, BRAF, and MSI/MMR.²

Results revealed that 51.6% of patients completed testing for all four biomarkers.² KRAS was tested in 81.6% of patients, NRAS in 60.5%, BRAF in 63.7%, and MSI/MMR in 73.4%.² Factors influencing testing rates included age, ethnicity, race, socioeconomic status, practice setting, and year of diagnosis, with younger, non-Hispanic white patients from higher socioeconomic backgrounds and academic settings, respectively, showing higher testing rates.² Despite increased testing rates over time, huge disparities persist, particularly among older vs younger, Hispanic or Latino vs non-Hispanic or Latino, and black vs white patients.² These findings underscore the need for equitable access to biomarker testing to ensure all patients benefit from precision oncology advances.²

AI-Based Model for Recurrence Prediction in T1 Low Rectal Cancer

Another notable study introduced an artificial intelligence (AI)-based model designed to predict recurrence risk in patients with T1 low rectal cancer post-local excision.³ Current guidelines recommend additional surgery for patients exhibiting high-risk features, despite a relatively low recurrence rate of approximately 7%.³ Such surgeries could adversely affect the anal function and overall quality of life.^3,4 To address this, researchers developed a convolutional neural network model using haematoxylin and eosin (H&E)-stained whole slide images from 507 patients treated between 2005 and 2015 across four hospitals.³ The model did not take into account any clinical features.³

The AI model demonstrated good discrimination and calibration, achieving a 5-year recurrence-free survival area under the receiver operating characteristic curve of 0.90, a sensitivity of 0.91, and a specificity of 0.82.³ The AI with a high level of accurately predicting recurrence reduced unnecessary additional surgeries by 25.7% compared to current guidelines-directed surgery.³ By providing a non-invasive and precise risk assessment tool, this model represents significant progress in personalised cancer care, improving clinical decision-making.³

References

Chakrabarti S, et al. J Clin Oncol. 2024;42(3_suppl):736.

Robinson HR, et al. J Clin Oncol. 2024;42(3_suppl):29.

Huang C, et al. J Clin Oncol. 2024;42(3_suppl):24.

Degiuli M, et al. Colorectal Dis. 2022;24(3):264–76.

Disclaimer

This article is not medical advice. Patients should seek personal assessment by a licenced specialist. Physicians are recommended to read the full publication(s) as cited in the article before making medical decisions. This article does not supersede nor replace the published article(s).