Artificial intelligence for predictive biomarker discovery in immuno-oncology: a systematic review.

BACKGROUND: The widespread use of immune checkpoint inhibitors (ICIs) has revolutionised treatment of multiple cancer types. However, selecting patients who may benefit from ICI remains challenging. Artificial intelligence (AI) approaches allow exploitation of high-dimension oncological data in research and development of precision immuno-oncology. MATERIALS AND METHODS: We conducted a systematic literature review of peer-reviewed original articles studying the ICI efficacy prediction in cancer patients across five data modalities: genomics (including genomics, transcriptomics, and epigenomics), radiomics, digital pathology (pathomics), and real-world and multimodality data. RESULTS: A total of 90 studies were included in this systematic review, with 80% published in 2021-2022. Among them, 37 studies included genomic, 20 radiomic, 8 pathomic, 20 real-world, and 5 multimodal data. Standard machine learning (ML) methods were used in 72% of studies, deep learning (DL) methods in 22%, and both in 6%. The most frequently studied cancer type was non-small-cell lung cancer (36%), followed by melanoma (16%), while 25% included pan-cancer studies. No prospective study design incorporated AI-based methodologies from the outset; rather, all implemented AI as a post hoc analysis. Novel biomarkers for ICI in radiomics and pathomics were identified using AI approaches, and molecular biomarkers have expanded past genomics into transcriptomics and epigenomics. Finally, complex algorithms and new types of AI-based markers, such as meta-biomarkers, are emerging by integrating multimodal/multi-omics data. CONCLUSION: AI-based methods have expanded the horizon for biomarker discovery, demonstrating the power of integrating multimodal data from existing datasets to discover new meta-biomarkers. While most of the included studies showed promise for AI-based prediction of benefit from immunotherapy, none provided high-level evidence for immediate practice change. A priori planned prospective trial designs are needed to cover all lifecycle steps of these software biomarkers, from development and validation to integration into clinical practice.

Upfront resection versus no resection of the primary tumor in patients with synchronous metastatic colorectal cancer: the randomized phase III CAIRO4 study conducted by the Dutch Colorectal Cancer Group and the Danish Colorectal Cancer Group.

BACKGROUND: Upfront primary tumor resection (PTR) has been associated with longer overall survival (OS) in patients with synchronous unresectable metastatic colorectal cancer (mCRC) in retrospective analyses. The aim of the CAIRO4 study was to investigate whether the addition of upfront PTR to systemic therapy resulted in a survival benefit in patients with synchronous mCRC without severe symptoms of their primary tumor. PATIENTS AND METHODS: This randomized phase III trial was conducted in 45 hospitals in The Netherlands and Denmark. Eligibility criteria included previously untreated mCRC, unresectable metastases, and no severe symptoms of the primary tumor. Patients were randomized (1 : 1) to upfront PTR followed by systemic therapy or systemic therapy without upfront PTR. Systemic therapy consisted of first-line fluoropyrimidine-based chemotherapy with bevacizumab in both arms. Primary endpoint was OS in the intention-to-treat population. The study was registered at ClinicalTrials.gov, NCT01606098. RESULTS: Between August 2012 and February 2021, 206 patients were randomized. In the intention-to-treat analysis, 204 patients were included (n = 103 without upfront PTR, n = 101 with upfront PTR) of whom 116 were men (57%) with median age of 65 years (interquartile range 59-71 years). Median follow-up was 69.4 months. Median OS in the arm without upfront PTR was 18.3 months (95% confidence interval 16.0-22.2 months) compared with 20.1 months (95% confidence interval 17.0-25.1 months) in the upfront PTR arm (P = 0.32). The number of grade 3-4 events was 71 (72%) in the arm without upfront PTR and 61 (65%) in the upfront PTR arm (P = 0.33). Three deaths (3%) possibly related to treatment were reported in the arm without upfront PTR and four (4%) in the upfront PTR arm. CONCLUSIONS: Addition of upfront PTR to palliative systemic therapy in patients with synchronous mCRC without severe symptoms of the primary tumor does not result in a survival benefit. This practice should no longer be considered standard of care.

An economical and highly adaptable optogenetics system for individual and population-level manipulation of Caenorhabditis elegans.

BACKGROUND: Optogenetics allows the experimental manipulation of excitable cells by a light stimulus without the need for technically challenging and invasive procedures. The high degree of spatial, temporal, and intensity control that can be achieved with a light stimulus, combined with cell type-specific expression of light-sensitive ion channels, enables highly specific and precise stimulation of excitable cells. Optogenetic tools have therefore revolutionized the study of neuronal circuits in a number of models, including Caenorhabditis elegans. Despite the existence of several optogenetic systems that allow spatial and temporal photoactivation of light-sensitive actuators in C. elegans, their high costs and low flexibility have limited wide access to optogenetics. Here, we developed an inexpensive, easy-to-build, modular, and adjustable optogenetics device for use on different microscopes and worm trackers, which we called the OptoArm. RESULTS: The OptoArm allows for single- and multiple-worm illumination and is adaptable in terms of light intensity, lighting profiles, and light color. We demonstrate OptoArm's power in a population-based multi-parameter study on the contributions of motor circuit cells to age-related motility decline. We found that individual components of the neuromuscular system display different rates of age-dependent deterioration. The functional decline of cholinergic neurons mirrors motor decline, while GABAergic neurons and muscle cells are relatively age-resilient, suggesting that rate-limiting cells exist and determine neuronal circuit ageing. CONCLUSION: We have assembled an economical, reliable, and highly adaptable optogenetics system which can be deployed to address diverse biological questions. We provide a detailed description of the construction as well as technical and biological validation of our set-up. Importantly, use of the OptoArm is not limited to C. elegans and may benefit studies in multiple model organisms, making optogenetics more accessible to the broader research community.

[Identification of differentiated dysplasia improves prediction of oral leukoplakia at increased risk of malignant progression]. / Onderkenning van gedifferentieerde dysplasie voor het voorspellen van de kans op maligne ontaarding van orale leukoplakie.

Oral leukoplakia is the most common potentially malignant disorder of the oral mucosa with a rate of malignant transformation into oral squamous cell carcinoma of 1-2% annually. The presence or absence of dysplasia as defined by the WHO is an important histological marker for malignant transformation risk assessment, but is not sufficiently accurate for patient stratification. We investigated whether identifying differentiated dysplasia contributes to oral leukoplakia malignant transformation risk assessment. We investigated whether classic or differentiated dysplasia were present in 84 oral leukoplakias. In 25 of these patients a squamous cell carcinoma developed during follow-up. Risk of malignant progression of oral leukoplakia increased from 3.3 (HR, p = 0.002) when only classic dysplasia was considered to 7.4 (HR, p = 0.001) when both classic and differentiated dysplasia were combined. This study demonstrates that identifying differentiated dysplasia as a separate type of dysplasia is important for the prognosis and stratification of patients with oral leukoplakia.

Pathological Complete Response Following Different Neoadjuvant Treatment Strategies for Locally Advanced Rectal Cancer: A Systematic Review and Meta-analysis.

BACKGROUND: Pathological complete response (pCR) following neoadjuvant treatment for locally advanced rectal cancer (LARC) is associated with better survival, less local recurrence, and less distant failure. Furthermore, pCR indicates that the rectum may have been preserved. This meta-analysis gives an overview of available neoadjuvant treatment strategies for LARC and analyzes how these perform in achieving pCR as compared with the standard of care. METHODS: Pubmed, Embase, and Cochrane Central bibliographic databases were searched. Randomized controlled trials in which patients received neoadjuvant treatment for MRI-staged nonmetastatic resectable LARC were included. The primary outcome was pCR, defined as ypT0N0. A meta-analysis of studies comparing an intervention with standard fluoropyrimidine-based chemoradiation (CRT) was performed. RESULTS: Of the 17 articles included in the systematic review, 11 were used for the meta-analysis. Addition of oxaliplatin to fluoropyrimidine-based CRT resulted in significantly more pCR compared with fluoropyrimidine-based CRT only (OR 1.46), but at the expense of more ≥ grade 3 toxicity. Other treatment strategies, including consolidation/induction chemotherapy and short-course radiotherapy (SCRT), did not improve pCR rates. None of the included trials reported a benefit in local control or OS. Five-year DFS was significantly worse after SCRT-delay compared with CRT (59% vs. 75.1%, HR 1.93). CONCLUSIONS: All included trials fail to deliver high-level evidence to show an improvement in pCR compared with standard fluoropyrimidine-based CRT. The addition of oxaliplatin might result in more pCR but at the expense of more toxicity. Furthermore, this benefit does not translate into less local recurrence or improved survival.

Circulating tumor DNA guided adjuvant chemotherapy in stage II colon cancer (MEDOCC-CrEATE): study protocol for a trial within a cohort study.

BACKGROUND: Accurate detection of patients with minimal residual disease (MRD) after surgery for stage II colon cancer (CC) remains an urgent unmet clinical need to improve selection of patients who might benefit form adjuvant chemotherapy (ACT). Presence of circulating tumor DNA (ctDNA) is indicative for MRD and has high predictive value for recurrent disease. The MEDOCC-CrEATE trial investigates how many stage II CC patients with detectable ctDNA after surgery will accept ACT and whether ACT reduces the risk of recurrence in these patients. METHODS/DESIGN: MEDOCC-CrEATE follows the 'trial within cohorts' (TwiCs) design. Patients with colorectal cancer (CRC) are included in the Prospective Dutch ColoRectal Cancer cohort (PLCRC) and give informed consent for collection of clinical data, tissue and blood samples, and consent for future randomization. MEDOCC-CrEATE is a subcohort within PLCRC consisting of 1320 stage II CC patients without indication for ACT according to current guidelines, who are randomized 1:1 into an experimental and a control arm. In the experimental arm, post-surgery blood samples and tissue are analyzed for tissue-informed detection of plasma ctDNA, using the PGDx elio™ platform. Patients with detectable ctDNA will be offered ACT consisting of 8 cycles of capecitabine plus oxaliplatin while patients without detectable ctDNA and patients in the control group will standard follow-up according to guideline. The primary endpoint is the proportion of patients receiving ACT when ctDNA is detectable after resection. The main secondary outcome is 2-year recurrence rate (RR), but also includes 5-year RR, disease free survival, overall survival, time to recurrence, quality of life and cost-effectiveness. Data will be analyzed by intention to treat. DISCUSSION: The MEDOCC-CrEATE trial will provide insight into the willingness of stage II CC patients to be treated with ACT guided by ctDNA biomarker testing and whether ACT will prevent recurrences in a high-risk population. Use of the TwiCs design provides the opportunity to randomize patients before ctDNA measurement, avoiding ethical dilemmas of ctDNA status disclosure in the control group. TRIAL REGISTRATION: Netherlands Trial Register: NL6281/NTR6455 . Registered 18 May 2017, https://www.trialregister.nl/trial/6281.

Improving clinical management of colon cancer through CONNECTION, a nation-wide colon cancer registry and stratification effort (CONNECTION II trial): rationale and protocol of a single arm intervention study.

BACKGROUND: It is estimated that around 15-30% of patients with early stage colon cancer benefit from adjuvant chemotherapy. We are currently not capable of upfront selection of patients who benefit from chemotherapy, which indicates the need for additional predictive markers for response to chemotherapy. It has been shown that the consensus molecular subtypes (CMSs), defined by RNA-profiling, have prognostic and/or predictive value. Due to postoperative timing of chemotherapy in current guidelines, tumor response to chemotherapy per CMS is not known, which makes the differentiation between the prognostic and predictive value impossible. Therefore, we propose to assess the tumor response per CMS in the neoadjuvant chemotherapy setting. This will provide us with clear data on the predictive value for chemotherapy response of the CMSs. METHODS: In this prospective, single arm, multicenter intervention study, 262 patients with resectable microsatellite stable cT3-4NxM0 colon cancer will be treated with two courses of neoadjuvant and two courses of adjuvant capecitabine and oxaliplatin. The primary endpoint is the pathological tumor response to neoadjuvant chemotherapy per CMS. Secondary endpoints are radiological tumor response, the prognostic value of these responses for recurrence free survival and overall survival and the differences in CMS classification of the same tumor before and after neoadjuvant chemotherapy. The study is scheduled to be performed in 8-10 Dutch hospitals. The first patient was included in February 2020. DISCUSSION: Patient selection for adjuvant chemotherapy in early stage colon cancer is far from optimal. The CMS classification is a promising new biomarker, but a solid chemotherapy response assessment per subtype is lacking. In this study we will investigate whether CMS classification can be of added value in clinical decision making by analyzing the predictive value for chemotherapy response. This study can provide the results necessary to proceed to future studies in which (neo) adjuvant chemotherapy may be withhold in patients with a specific CMS subtype, who show no benefit from chemotherapy and for whom possible new treatments can be investigated. TRIAL REGISTRATION: This study has been registered in the Netherlands Trial Register (NL8177) at 11-26-2019, https://www.trialregister.nl/trial/8177 . The study has been approved by the medical ethics committee Utrecht (MEC18/712).

Feasibility and effectiveness of trifluridine/tipiracil in metastatic colorectal cancer: real-life data from The Netherlands.

BACKGROUND: The RECOURSE trial showed clinical efficacy for trifluridine/tipiracil for refractory metastatic colorectal cancer patients. We assessed the feasibility and effectiveness of trifluridine/tipiracil in daily clinical practice in The Netherlands. METHODS: Medical records of patients from 17 centers treated in the trifluridine/tipiracil compassionate use program were reviewed and checked for RECOURSE eligibility criteria. Baseline characteristics, safety, and survival times were compared, and prespecified baseline characteristics were tested in multivariate analyses for prognostic significance on overall survival (OS). RESULTS: A total of 136 patients with a median age of 62 years were analyzed. Forty-three patients (32%) did not meet the RECOURSE eligibility criteria for not having received all prior standard treatments (n = 35, 26%) and/or ECOG performance status (PS) 2 (n = 12, 9%). The most common grade ≥3 toxicities were neutropenia (n = 44, 32%), leukopenia (n = 8, 6%), anemia (n = 7, 5%), and fatigue (n = 7, 5%). Median progression-free survival (PFS) and median OS were 2.1 (95% CI, 1.8-2.3) and 5.4 months (95% CI, 4.0-6.9), respectively. Patients with ECOG PS 2 had a worse median OS (3.2 months) compared to patients with ECOG PS 0-1 (5.9 months). ECOG PS, KRAS-mutation status, white blood cell count, serum lactate dehydrogenase, and alkaline phosphatase were prognostic factors for OS. CONCLUSIONS: Our data show that treatment with trifluridine/tipiracil in daily clinical practice is feasible and safe. Differences in patient characteristics between our population and the RECOURSE study population should be taken into account in the interpretation of survival data. Our results argue against the use of trifluridine/tipiracil in patients with ECOG PS 2. FUNDING: Johannes J.M. Kwakman received an unrestricted research grant from Servier.

Short versus conventional hydration for prevention of kidney injury during pre-TAVI computed tomography angiography.

BACKGROUND: Computed tomography angiography (CTA) is required in the work-up for transcatheter aortic valve implantation (TAVI). However, CTA may cause contrast-induced acute kidney injury (CI-AKI). We hypothesised that a short (1 h, 3 ml/kg/h sodium bicarbonate) hydration protocol is not inferior to conventional (24 h, 1 ml/kg/h saline) hydration in avoiding a decline in renal function in patients with impaired renal function. METHODS AND RESULTS: Single-centre randomised non-inferiority trial in patients with impaired renal function who underwent pre-TAVI CTA. Patients were randomised on a 1:1 ratio to short hydration (SHORT; 1 h sodium bicarbonate, 3 ml/kg/h) or conventional hydration (CONV; 24 h saline, 1 ml/kg/h). Outcomes included percentage change in serum creatinine until 2-6 days after CTA with a non-inferiority margin of 10% and an increase on the Borg dyspnoea scale ≥1 point. Seventy-four patients were included. Increase in creatinine was 6 µmol/l (95% CI 2.5-9.3) in the SHORT versus 2 µmol/l (95% CI-1.4 to 6.3) in the CONV arm (p = 0.167). The percentage change was 4.6% (95% CI 2.0-7.3%) in the SHORT arm versus 2.5% (95% CI: 0.8 to 5.8%) in the CONV arm. The difference in percentage increase in creatinine between the two arms was 2.1% (95% CI: 2.0-6.2%; p-value non-inferiority: <0.001). CI-AKI and a need for dialysis were not observed. An increase of ≥1 point on the Borg scale (dyspnoea scale ranging from 1 (lowest) to 10 (highest)) was seen in 1 patient in the SHORT arm versus 5 patients in the CONV arm (2.9% vs 16.1%, p = 0.091). CONCLUSION: For patients with impaired renal function undergoing pre-TAVI CTA, a short 1­h, low-volume hydration protocol with sodium bicarbonate is not inferior to conventional 24-h, high-volume saline hydration.