Artificial Intelligence for Help Non-Small Cell Lung Cancer: Measure Cancer Biology and Treatment Response Via Imaging

Lung cancer is the leading cause of cancer-related death worldwide, accounting for an estimated 2.1 million deaths in 2018. About 80%-85% of lung cancers are NSCLC. The main subtypes of NSCLC are adenocarcinoma, squamous cell carcinoma, and large cell carcinoma. These subtypes, which start from different types of lung cells, are grouped together as NSCLC. For these reasons, the choice of NSCLC for this trial is the most relevant, as there is a clear societal burden / clinical need that will be addressed in SALMON.

A pressing need exists to uncover the genetic and molecular composition of tumors, as this information would accelerate the development of more effective cancer therapies. Since tumors differ in their biological makeup, treatments can now often be tailored towards individual patients, in a strategy termed personalized medicine. NSCLC perfectly illustrates this paradigm, with treatments either targeted at well-defined oncogenic pathways (epidermal growth factor receptor (EGFR) mutations, Anaplastic lymphoma kinase (ALK) \& receptor tyrosine kinase 1 (ROS1) gene rearrangements, B-Rapidly Accelerated Fibrosarcoma gene (BRAF) mutations), and for so-called "wild-type" NSCLC, immunotherapy (with or without chemotherapy). The recent development of different types of immunotherapy has led to promising advances in the treatment of patients with NSCLC in advanced or metastatic disease.

Medical imaging plays a pivotal role in the assessment of tumors, including lung cancer, as it provides non-invasively the features to identify, characterize and stage the local tumor and its overall metastatic burden. The information provided by imaging can be still improved with a thorough post-processing analysis of the images, with "radiomics" being the most sophisticated tool for quantitative imaging. Through radiomics the imaging data obtained can be linked to specific biological properties of the tumor, acquired either from tissue or liquid biopsies. However, the knowledge on this is fragmented to date.

The first objective of this trial is to generate a radiomics atlas (rATLAS) that provides a broader link between biological descriptors and medical imaging features in NSCLC, to eventually identify oncologic pathways through medical images and quantify the overall burden of specific markers for targeted therapies. In particular, imaging biomarkers that can predict immunosensitivity and more accurately predict the prognosis than the existing ones will be assessed on the medical images and compared to tissue and liquid biopsies.

Apart from NSCLC diagnosis and staging, imaging plays a key role also in the evaluation of patient's follow-up, representing the gold standard for the assessment of response to therapy. However, the evaluation of medical images is amenable to radiologist's interpretation and many attempts have been made to face this issue in the last decades. The RECIST is a one-dimensional measure (shortest/longest diameter in the plane of measurement) created in 2000, updated in 2009 (RECIST 1.1), with a branch for immunotherapy finalized in 2017 (iRECIST). RECIST, in all its different declination, is used to assess whether a tumor in cancer patients is progressing, regressing or did not change before and after some event, such as therapy. RECIST criteria are recognized both in the EU and in US. Although representing an international attempt to overcome the subjective evaluation of tumor response to cancer treatment, biases of measurements remain as it relies on human assessments. In 1976 Moertel and Hanley acknowledged that "the culmination of most experimental therapeutic trials for solid tumors occurs when a [physician] places a ruler or caliper over a lump and attempts to estimate its size," and with this measurement comes the inevitable component of human error. In 2021 this issue is still present in clinical practice, but in the last years Artificial Intelligence (AI) has demonstrated remarkable progress in image-recognition tasks in research.

Thus, the investigators propose to eliminate the shortcomings of measurement biases with the second objective of our trial that is focused on the validation of a new automated RECIST (aRECIST) workflow for the assessment of therapy response.

A signed and dated informed consent form will be obtained from each patient before any screening procedures are conducted. The participants will be prospectively enrolled to rATLAS only, to both rATLAS and aRECIST, or considered as screen failure. If enrolled then a unique study-specific subject number shall be assigned, Informed consent form will be signed and the following evaluations will be included:

The following data will be obtained for each participant from medical files and records:

• Demographic data (Age, Gender, Race, country);
• Anthropometric data (Weight in kg, Height in cm, body mass index (BMI) in kg/m^2);
• Medical history (comorbidities, oncology-related history, Chronic obstructive pulmonary disease (COPD), Interstitial lung disease (ILD), Pulmonary vascular disease, history related to NSCLC);
• Substance use (cigarettes/e-cigarettes/alcohol), duration and dose/frequency [smoking required, others optional];
• Concomitant treatment only linked to immunotherapy, chemotherapy, radiotherapy, and targeted therapy (subtype, drug name, line of treatment (first/second/third line), reason for discontinuation (completed regimen/completed response/progression disease/toxicity/other);
• Eastern Cooperative Oncology Group (ECOG) performance status (0 to 5);
• Histological diagnosis (adenocarcinoma, squamous cell carcinoma, large cell carcinoma, adenosquamous cell carcinoma, other);
• Laboratoristic data: Hematology (hemoglobin, hematocrit, leukocytes, neutrophils, eosinophils, basophils, lymphocytes, monocytes, platelets) and Chemistry (sodium, bicarbonate, chloride, calcium, albumin, magnesium, phosphate, blood urea nitrogen, protein, urea, potassium, creatinine, estimated glomerular filtration rate, bilirubin, direct bilirubin, aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase, lactate dehydrogenase (LDH), carcinoembryonic antigen (CEA), squamous cell carcinoma (SCC)).

From clinical examination and imaging (and multidisciplinary discussion report if available) the following data will be recorded per participant: Clinical Tumor Staging (cTNM) upon initial diagnosis.

From Radiology department the following data will be recorded per participant: The staging computed tomography (CT) scan data already performed at the site per standard of care (SoC) will be pseudo-anonymized and sent to the central reader, along with Positron Emission Tomography (PET)/CT and brain magnetic resonance imaging (MRI) scans obtained per patient staging, when available.

Then, depending on group (aRECIST or rATLAS), the following steps will also be considered:

rATLAS: Diagnostic samples of tissue biopsy (ideally 1 block or 25 slides - minimum required: 10 slides) performed at the participating site per standard of care will be shipped to the central laboratory for testing on oncogenic biomarkers. After this baseline visit all participants in rATLAS may start their treatment as per SoC. rATLAS study ends after this baseline and enrolment visit.

aRECIST: A liquid biopsy will be obtained (blood sample, 1 tube of 10 ml) and shipped within 2 days to the central laboratory for oncogenic biomarker testing. In order not to influence therapeutic decision making, biomarker results will be shared with the sites only after the decision concerning NSCLC therapy has been made. Diagnostic samples of tissue biopsy which have been obtained for ATLAS (ideally 1 block or 25 slides - minimum required: 10 slides) performed at the participating site per standard of care will be shipped to the central laboratory for oncogenetic biomarker testing. In those cases where standard tissue biopsy is not obtained, fine needle biopsy and cells obtained per SoC by fine needle aspiration could be analyzed and send to the central laboratory following specific procedures. Enough material should be available to be analyzed, otherwise the genetic testing would not be possible.

After baseline all participants of the aRECIST group will undergo standard chemotherapy and/or immunotherapy and will be followed up for a maximum of 2 years to investigate a possible treatment response (manual local \& central and automated RECIST) by imaging parameters and liquid biopsy. These participants will need to return to the clinic for liquid biopsies (2 tubes of 10 mL) at 3, 6, 12 and 24 months within 5 days (before or after) of their imaging follow-up as per SoC. It is the site staff that will be responsible for providing a copy of all relevant post-treatment CT-scans (and PET/CT scan if available) as taken per SoC together with local RECIST scoring, NSCLC treatment(s) information (class, dose, duration) and NSCLC-related events. The data will be collected at time-points as per standard of care (at 3, 6, 12, and 24 months approximately). The participant's participation to aRECIST will end 2 years after baseline or earlier in case of death or any event that in the opinion of the investigator requires an early termination.

In particular, the site staff will be responsible to provide a liquid biopsy (2 tubes of 10ml each) and to share the following data at 3 (visit1), 6 (visit 2), 12 (visit 3), and 24 (visit 4) months):

• A copy of all relevant post-treatment CT scans - and PET/CT scans if available - as taken per SoC
• Detailed description of CT acquisition for each participant, including contrast media injection characteristics
• Local reading of treatment response by using RECIST 1.1 criteria (manual local reading), along with the time needed for evaluating each participant,
• A liquid biopsy (2 tubes of 10 ml each) within 5 days of the CT scan performed for participant follow up
• NSCLC treatment(s) information (class, dose, duration, specific timeframes)
• NSCLC-related events.

Any rATLAS participant are considered as having completed the study after performing the initial baseline visit (Day 0), except for those also participating to aRECIST. Participants of aRECIST are considered as having completed the study after the 2-year follow-up period, or earlier, in case of death or any event that in the opinion of the investigator requires an early termination.