ESMO 2024 Highlights

• New data have been presented for Trastuzumab-deruxtecan (T-DxT, Enhertu) already approved for HER2-positive breast cancer. The DESTINY-Breast12 study demonstrated durable intercranial activity in patients with brain metastases (Abstract LBA18). However, pathological assessment of HER2-low levels in order to select patients for T-DxT treatment remains a challenge. The DESTINY-Breast06 study examined the concordance of HER2 assessment between local and central labs and reported it to be approximately 78%. Out of 349 patients scored as HER2 null locally, 64% were found to be HER2-low or HER2-ultralow by central testing (Abstract LBA21). These results underscore the urgency to improve biomarker testing in order to select all eligible patients for T-DxT treatment.
• Claudin-targeted antibody–drug conjugates (ADCs)seem to be a promising emerging treatment option for solid tumors. Two ADCs targeting claudin 18.2 (SHR-A1904, Abstract 609O and XNW27011, Abstract 651P)demonstrated manageable safety profiles and promising efficacy results in early trials in gastric, gastroesophageal and gastrointestinal tumors. Treatment with claudin-6 targeting (CLDN-6) ADC DS-9606ademonstrated manageable and tolerable safety in solid tumors (Abstract 610O).Encouraging early results have also been reported for TORL-1-23 targeting CLDN-6in solid tumors (Abstract 721MO).Although an unselected patient population was enrolled in the phase I study, CLDN-6 positive ovarian patients in particular responded well to treatment with an objective response rate (ORR) of 50% compared to 26% in overall population. The Phase II trial will include CLDN-6 positive patients, emphasizing the importance of biomarker testing and selecting the right patient for the right treatment. 
• PRMT5 is an emerging target in patients with MTAP deletion. An early-phase trial with AMG-193 has shown acceptable safety profiles and promising anti-tumor activity across different solid tumors, with an ORR of 11.8%, 8.7% and 10.5% in patients with non-small cell lung cancer (NSCLC), pancreatic ductal adenocarcinoma and biliary tract cancer respectively (Abstract 604O).
• Precision oncology is making advancements across different indications, with biomarker testing as an integral part of patient care.
  
  To name a few:
  
  
  • Results from the phase II INSPIRE study demonstrated promising efficacy of nivolumab with ipilimumab (Opdivo-Yervoy)combination treatment in deficient mismatch repair (dMMR) prostate cancer (Abstract 642TiP).
  • A new PD-L1-directed ADC, SGN-PDL1V, showed encouraging early data in NSCLC and HNSCC PD-L1-positive tumors (Abstract 607O).
  • KRAS G12Cremains target of interest. Data from theKRYSTAL-12 trial reconfirmed adagrasib (Krazati) as an efficacious treatment option for KRAS-mutated NSCLC patients and suggested intracranial efficacy inpatients with brain metastases (although patient numbers were low)(Abstract LBA57). 
  • Sotorasib (Lumakras), already approved for NSCLC, is showing efficacy in other indications as well. In combination with anti-EGFR monoclonal antibody panitumumab (Vectibix) and chemo in first-line KRAS G12C-mutant metastatic colorectal cancer showed high response rate (tumor shrinkage in more than 75% of patients) according to data from the phase I Code BreaK 101 study (Abstract 505O). 
  • The KEYNOTE-811 trial demonstrated improved overall survival (OS) with first-line pembrolizumab in combination with trastuzumab and chemo vs. placebo (20months vs.16.8 months) in patients with treatment-naive unresectable, HER2-positive metastatic gastric/gastro-esophageal junction carcinoma (Abstract 1400O). 
  • Promising efficacy results were displayed from early-phase studies for new tyrosine kinase inhibitors (TKIs) in ALK-positive (NVL-655, ALKOVE study, Abstract 1253O)and ROS1-positive (NVL-520, ARROS-1 study, Abstract 1256MO) NSCLC, which could be potential future therapeutic options for the patients who developed resistance to previous lines of treatment.
  • ASP3082, a KRAS G12D selective protein degrader, demonstrated an acceptable safety profile, and the promising ORR of 23% in NSCLC and 18.5% in pancreatic KRAS G12D mutated cancers (Abstract 608O).
    
    Looking to the future
• A tumor-agnostic approach seems to be the way to move forward. This was supported by results from the ROME trial(Abstract LBA7), a randomized, multi-basket phase II study where 400 pre-treated patients with advanced or metastatic solid tumors across different indications with detected actionable genetic alterations were treated with different targeted treatments or immune checkpoint inhibitors, ultimately reporting a significantly improved ORR (17.0% versus 9.5%, p=0.027) and median progression-free survival (PFS; 3.7 months vs. 2.8 months, p<0.0001) vs. standard-of-care. The role of a molecular tumor board was critical for the assessment of detected genetic alterations and treatment selection. In an effort to support future tumor-agnostic trials, ESMO created the Tumor-Agnostic Classifier and Screener (ETAC-S) tool, providing minimal requirements for assessing and defining the tumor-agnostic potential of molecularly guided treatment options.
• Artificial intelligence (AI) is becoming an integral part of precision oncology and medical care, although clinical validation and regulatory requirements remain challenging for AI-driven medical devices. GigaPath has been discussed at the ESMO presidential symposium as a promising new AI tool for digital pathology, with improved mutation prediction and cancer subtyping that can be utilized for clinical diagnostics and to support decision making (Abstract 1942O,Xu et al, Nature, 2024). Moreover, the AI-driven genomic test TNBC-DX demonstrated potential in a retrospective validation study, enabling identification of early-stage, triple-negative breast cancer patients likely to benefit from neoadjuvant taxane-carboplatin chemotherapy (Abstract 237MO).In order to support further progress in the field, ESMO is organizing its first AI and Digital Oncology congress in 2025.
• Liquid biopsy testing and ctDNA detection are being extensively studied across different cancer types, especially in early-stage disease, for risk stratification and minimal residual disease (MRD)detection, as well as for guiding treatment decisions. Using Natera’s Signatera assay, the interim analysis of the CIRCULATE-Japan GALAXY observational study demonstrated that ctDNA-based MRD detection can accurately predict risk of recurrence and potential benefit from adjuvant chemotherapy in resectable colorectal cancer (Abstract 553P, 554P, 555P, 558P, Nakamura et al, Nat. Med., 2024). The predictive value of ctDNA has also been demonstrated in early-stage lung cancer (Abstract 1226P; IMpower010 study, Abstract 1211P;ctDNA-Lung-Detect trial, Abstract 1236P). However, test sensitivity and high false negative rates remain suboptimal, and improvements are urgently needed. In an effort to bypass this obstacle, a group from Stanford university is developing PhaseED-Seq, personalized cancer profiling by deep sequencing, and presented impressive data on improved limit of detection and sensitivity they were able to achieve with this approach in breast cancer (Abstract 293P).These kinds of innovations can be utilized to optimize early MRD detection. 

Conclusions:

• Upfront biomarker testing, as well as accurate biomarker assessment, is critical for selecting eligible patients for appropriate treatments. More optimization and education in the field is needed to achieve seamless testing across different indications.
• A tumor-agnostic approach is likely to accelerate the generation of clinical data and expand the existing pool of patients eligible for treatments based on biomarker selection. It is important to incorporate this approach into a new clinical trial framework where applicable.
• Growing data are demonstrating the importance of liquid biopsy and ctDNA monitoring in the early onset of cancer, in order to detect the disease when patients have more chances of getting cured, and to prevent overtreating patients with unnecessary toxic treatments. Technological improvements of sensitivity and subsequent clinical validation are critical to ensure patients get properly tested.
• AI tools are being increasingly utilized in medical care, from digital pathology to various steps of clinical trials. It is necessary to improve the quality of data and clinical validation protocols as well as to adjust regulatory frameworks to enable further development in the field.

If you missed us at ESMO, contact us here to talk to our team about our unique range of solutions to support end-to-end commercialization.