BackgroundMultidisciplinary team (MDT) meetings are used to optimise expert decision-making about treatment options, but such expertise is not digitally transferable between centres. To help standardise medical decision-making, we developed a machine learning model designed to predict MDT decisions about adjuvant breast cancer treatments.MethodsWe analysed MDT decisions regarding adjuvant systemic therapy for 1065 breast cancer cases over eight years. Machine learning classifiers with and without bootstrap aggregation were correlated with MDT decisions (recommended, not recommended, or discussable) regarding adjuvant cytotoxic, endocrine and biologic/targeted therapies, then tested for predictability using stratified ten-fold cross-validations. The predictions so derived were duly compared with those based on published (ESMO and NCCN) cancer guidelines.ResultsMachine learning more accurately predicted adjuvant chemotherapy MDT decisions than did simple application of guidelines. No differences were found between MDT- vs. ESMO/NCCN- based decisions to prescribe either adjuvant endocrine (97%, p = 0.44/0.74) or biologic/targeted therapies (98%, p = 0.82/0.59). In contrast, significant discrepancies were evident between MDT- and guideline-based decisions to prescribe chemotherapy (87%, p < 0.01, representing 43% and 53% variations from ESMO/NCCN guidelines, respectively). Using ten-fold cross-validation, the best classifiers achieved areas under the receiver operating characteristic curve (AUC) of 0.940 for chemotherapy (95% C.I., 0.922—0.958), 0.899 for the endocrine therapy (95% C.I., 0.880—0.918), and 0.977 for trastuzumab therapy (95% C.I., 0.955—0.999) respectively. Overall, bootstrap aggregated classifiers performed better among all evaluated machine learning models.ConclusionsA machine learning approach based on clinicopathologic characteristics can predict MDT decisions about adjuvant breast cancer drug therapies. The discrepancy between MDT- and guideline-based decisions regarding adjuvant chemotherapy implies that certain non-clincopathologic criteria, such as patient preference and resource availability, are factored into clinical decision-making by local experts but not captured by guidelines.Electronic supplementary materialThe online version of this article (doi:10.1186/s12885-016-2972-z) contains supplementary material, which is available to authorized users.

Vast amounts of clinically relevant text-based variables lie undiscovered and unexploited in electronic medical records (EMR). To exploit this untapped resource, and thus facilitate the discovery of informative covariates from unstructured clinical narratives, we have built a novel computational pipeline termed Text-based Exploratory Pattern Analyser for Prognosticator and Associator discovery (TEPAPA). This pipeline combines semantic-free natural language processing (NLP), regular expression induction, and statistical association testing to identify conserved text patterns associated with outcome variables of clinical interest. When we applied TEPAPA to a cohort of head and neck squamous cell carcinoma patients, plausible concepts known to be correlated with human papilloma virus (HPV) status were identified from the EMR text, including site of primary disease, tumour stage, pathologic characteristics, and treatment modalities. Similarly, correlates of other variables (including gender, nodal status, recurrent disease, smoking and alcohol status) were also reliably recovered. Using highly-associated patterns as covariates, a patient’s HPV status was classifiable using a bootstrap analysis with a mean area under the ROC curve of 0.861, suggesting its predictive utility in supporting EMR-based phenotyping tasks. These data support using this integrative approach to efficiently identify disease-associated factors from unstructured EMR narratives, and thus to efficiently generate testable hypotheses.

Aim: To review the expected increasing demand for cancer services among low and middle-income countries (LMICs) in the Asia-Pacific (APAC), and to describe ways in which Australia and New Zealand (ANZ) can provide support to improve cancer outcomes in our region. Methods:We first review the current and projected incidence of cancer within the APAC between 2018 and 2040, and the estimated demand for chemotherapy, radiotherapy and surgery. We then explore potential ways in which ANZ can increase regional collaborations to improve cancer outcomes. Results:We identify 6 ways that ANZ can collaborate with LMICs to improve cancer care in the APAC through the ANZ Regional Oncology Collaboration Strategy: 1. Increasing education and institutional collaborations in the APAC region through incountry training, twinning partnerships, observerships and formalised training programs in order to increase cancer care quality and capacity. 2. Promoting and assisting in the establishment and maintenance of population-based cancer registries in LMICs. 3. Increasing research capacity in LMICs through collaboration and promoting high quality global oncology research within ANZ. 4. Engaging and training Australian and New Zealand clinicians in global oncology, increasing awareness of this important career path, and increasing health policy engagement. 5. Increasing web-based endeavours through virtual tumour boards, web-based advocacy platforms and web-based teaching programs. 6. Continuing to leverage for funding through professional bodies, government, industry, not-for-profit organisations and local hospital funds. Conclusion:We propose the creation of an Australian and New Zealand Interest Group to provide formalised and sustained collaboration between researchers, clinicians and stakeholders.

BackgroundPayments to medical oncologists and clinical haematologists can negatively affect prescribing practice, but the extent of payments to these specialists is unknown in Australia. AimsWe aimed to analyse the extent of payments from the pharmaceutical industry to Australian cancer physicians as reported during the first collated period of the Disclosure Australia website. MethodsWe performed a retrospective, cross-sectional analysis of payments made from November 2018 to April 2019, using a file downloaded from the Disclosure Australia website. We checked the names of listed medical practitioners against Medical Board of Australia records to assign specialties. The number of medical oncologists, clinical haematologists, other specialist physicians and non-specialist physician medical practitioners was calculated, along with the payments to each of these groups.

PURPOSE: Interactions between cancer physicians and the pharmaceutical industry may create conflicts of interest that can adversely affect patient care. We aimed to survey cancer physicians regarding their attitudes toward and interactions with industry. METHODS: We surveyed Australian cancer physicians between December 2020 and February 2021, questioning how often they interacted with industry and their attitudes toward this. We also assessed factors associated with accepting payments from industry and the amount received, and opinions on policies and industry influence. We used logistic and linear regression to examine links between attitudes and behaviors. RESULTS: There were 116 responses (94 complete). Almost half (n = 53 of 115, 46.1%) felt that there was a positive relationship between cancer physicians and industry. Most (n = 79 of 104, 76.0%) interacted with industry at least once a month, and 67.7% (n = 63 of 93) had received nonresearch payments from industry previously, with a median value of 2,000 Australian dollars over 1 year. Most respondents believed that interactions could influence prescribing while simultaneously denying influence on their own prescribing (n = 66 of 94, 70.2%). Those who judged general sales representative interactions (odds ratio [OR] 9.37 [95% CI, 1.05 to 83.41], P = .045) or clinician sponsorship (OR 3.22 [95% CI, 1.01 to 10.30], P = .049) to be more acceptable also met with sales representatives more frequently. Physicians were more likely to accept industry payments when they deemed sponsorship of clinicians for conferences (OR 10.55 [95% CI, 2.33 to 47.89], P = .002) or honoraria for advisory board membership more acceptable (OR 3.91 [95% CI, 1.04 to 14.74], P = .04) or when they had higher belief in industry influence over own prescribing (OR 25.51 [95% CI, 2.70 to 241.45], P = .005). CONCLUSION: Australian cancer physicians interact with industry frequently, and those who feel positive about these interactions are likely to do so more often. More research is needed to understand the motivations behind these interactions.

A quarter of patients 80 years and older received first-line palliative chemotherapy. Despite most receiving a modified dose, one third were hospitalised during treatment. These findings highlight the need for careful clinical assessment and selection of older cancer patients for chemotherapy.

Background: No previous review has assessed the extent and effect of industry interactions on medical oncologists and haematologists specifically.Methods: A systematic review investigated interactions with the pharmaceutical industry and how these might affect the clinical practice, knowledge and beliefs of cancer physicians. MEDLINE, Embase, PsycINFO and Web of Science Core Collection databases were searched from inception to February, 2021.Results: Twenty-nine cross-sectional and two cohort studies met the inclusion criteria. These were classified into three categories of investigation: 1. Extent of exposure to industry for cancer physicians as whole (n=11); 2. Financial ties among influential cancer physicians specifically (n=11), and; 3. Associations between industry exposure and prescribing (n=9).Cancer physicians frequently receive payments from or maintain financial ties with industry, at a prevalence of up to 63% in the United States (US) and 70.6% in Japan. Among influential clinicians, 86% of US and 78% of Japanese oncology guidelines authors receive payments. Payments were associated with either a neutral or negative influence on the quality of prescribing practice. Limited evidence suggests oncologists believe education by industry could lead to unconscious bias.Conclusions: There is substantial evidence of frequent relationships between cancer physicians and the pharmaceutical industry in a range of high income countries. More research is needed on clinical implications for patients and better management of these relationships.

No abstract