REVIEW ONCOLOGY

Artificial intelligence for the oncologist: hype, hubris or reality?

Keeping up with the rising amount of clinical data, guidelines and approvals of new antineoplastic drugs is a major challenge for every oncologist. Artificial intelligence promises to address this and to revolutionise health care and cancer treatment. What is the current state of artificial intelligence for the oncologist, and is it ready for prime time? In this article, the current, available tools of artificial intelligence are highlighted, which try to take a privileged role in the daily practice of the oncologist.

(BELG J MED ONCOL 2018;12(7)330–333)

Prognostic and predictive value of primary tumour location in metastatic colorectal cancer: will the side change our treatment?

Metastatic colorectal cancer is a heterogeneous disease. Tumours arising from different regions of the colon are clinically and molecularly distinct. The differing molecular characteristics translate into a differential clinical outcome with right-sided tumours displaying a worse prognosis compared to left-sided tumours. Besides the prognostic relevance of the primary tumour location, several retrospective analyses suggest that the primary tumour location may also be predictive of treatment benefit from targeted therapy with anti-EGFR and anti-VEGF directed agents in the first-line treatment of RAS wild-type metastatic colorectal cancer.

(BELG J MED ONCOL 2018;12(6):271–274)

ALK-inhibitors in NSCLC: a tabulated review of randomised phase III trials

Three inhibitors ALK-inhibitors have been assessed in six randomised phase III trials in patients with advanced anaplastic lymphoma kinase (ALK) positive Non Small-Cell Lung Cancer (NSCLC). Patient and tumour characteristics, efficacy, and safety results are summarised in Tables 1, 2, and 3, respectively. Progression-free survival was the primary endpoint in all trials. In first-line, alectinib was superior to crizotonib, and both crizotinib and ceritinib were superior to pemetrexed plus platinum. Crizotinib and ceritinib were superior to pemetrexed or docetaxel in previously treated patients.
(BELG J MED ONCOL 2018;12(6):275–278)

Emerging concepts in urothelial cancer

Treatment for urothelial cancer has undergone rapid change. Cisplatin based chemotherapy should be given in the neo-adjuvant setting in muscle invasive bladder cancer and could play a role in trimodality therapy when combined with surgery and radiotherapy. Genetic profiling has differentiated several subtypes of urothelial cancer, mimicking progress seen in breast cancer. Of these subtypes, p53 like tumours are less likely to respond to neo-adjuvant chemotherapy. In metastatic urothelial cancer, systemic immunotherapy (checkpoint inhibitors) has shown promising results in first line and second line patients. In a phase III trial, pembrolizumab, an anti-PD1 (programmed cell death 1) antibody, showed a survival benefit in second line metastatic urothelial cancer and should be the new standard of care. In patients who are cisplatin ineligible checkpoint can be used in first line, but no phase III data are available.

(BELG J MED ONCOL 2018;12(5):212–217)

Immuno-oncology in transitional cell Carcinoma

Immunotherapy has been used in the treatment of localised, high-risk transitional cell carcinoma. Bacillus of Calmette-Guérin therapy is a standard treatment for patients with non-invasive transitional cell carcinoma with bad prognostic factors (high-grade pTa; carcinoma in situ) and early stage invasive bladder cancer (pT1) after transurethral resection of the bladder. Recently, based on phase II trials, atezolizumab, an inhibitor of PD-L1, and nivolumab, an inhibitor of PD1, have been registered for the treatment of patients with metastatic transitional cell carcinoma progressing after a platinum-based chemotherapy for metastatic disease. Pembrolizumab, a monoclonal antibody against programmed death receptor-1 was registered based on a phase III trial in this setting resulting in a survival benefit compared to second-line chemotherapy. Predictive markers are being explored for a better patient selection for the treatment of transitional cell carcinoma.

(BELG J MED ONCOL 2018;12(5):218–222)

Targeting driver mutations in non-small-cell lung cancer, beyond the usual suspects (EGFR and ALK)

In the last decade, systemic treatment for non-small-cell lung cancer has undergone an unprecedented change because of new targeted therapies and the introduction of immunotherapy. Advances in the understanding of lung cancer biology have led to the discovery of several oncogenic driver genes and the development of drugs that target driver mutations, according to the strategy of ‘personalised therapy’. The bestknown alterations are epidermal growth factor mutations and anaplastic lymphoma kinase rearrangements, but the improvement in genomic technologies and the continuous research in this area have led to the identification of new druggable targets. This is a comprehensive overview focused on the development of targeted therapies and their mechanisms of action.

(BELG J MED ONCOL 2018;12(5):223–232)

Brain metastases: Systemic treatment approach in 2017

This review proposes to go through reasonable systemic therapy options in brain metastases, notably immune checkpoint inhibitors and oncogen-driven targeted therapies. We deliberately focus on drugs currently available in Belgium in clinical practice. In the large majority of cases, clinical trials – in particular registration trials – exclude patients with brain metastases. Therefore we have to deal with small size non-randomised phase II trials or retrospective analysis with the known caveats of highly selected patients and numerous biases.

(BELG J MED ONCOL 2018:12(3):96–102)