Cancer Treatment Decisions Research Articles

MO14-6 The impact of patient decision aids on the shared decision-making in cancer treatment

Shared decision-making (SDM) has been identified as a key element in patient-centered oncology care. Proper patient decision aids (PDAs) in the patient-professional interactions are necessary for the quality of SDM. High-quality PDAs assist patients in making informed values-based choices and promote patient-professional interactions at the same time. The study aimed to explore whether existing and emerging PDAs improved SDM and supported patient-professional interactions from the perspectives of oncology users.

Non-invasive prediction of microsatellite instability in colorectal cancer by a genetic algorithm-enhanced artificial neural network-based CT radiomics signature.

The stratification of microsatellite instability (MSI) status assists clinicians in making treatment decisions for colorectal cancer (CRC) patients. This study aimed to establish a CT-based radiomics signature to predict MSI status in patients with CRC. A total of 837 CRC patients who underwent preoperative enhanced CT and had available MSI status data were recruited from two hospitals. Radiomics features were extracted from segmented tumours, and a series of data balancing and feature selection strategies were used to select MSI-related features. Finally, an MSI-related radiomics signature was constructed using a genetic algorithm-enhanced artificial neural network model. Combined and clinical models were constructed using multivariate logistic regression analyses by integrating the clinical factors with or without the signature. A Kaplan-Meier survival analysis was conducted to explore the prognostic information of the signature in patients with CRC. Ten features were selected to construct a signature which showed robust performance in both the internal and external validation cohorts, with areas under the curves (AUC) of 0.788 and 0.775, respectively. The performance of the signature was comparable to that of the combined model (AUCs of 0.777 and 0.767, respectively) and it outperformed the clinical model constituting age and tumour location (AUCs of 0.768 and 0.623, respectively). Survival analysis demonstrated that the signature could stratify patients with stage II CRC according to prognosis (HR: 0.402, p = 0.029). This study built a robust radiomics signature for identifying the MSI status of CRC patients, which may assist individualised treatment decisions. • Our well-designed modelling strategies helped overcome the problem of data imbalance caused by the low incidence of MSI. • Genetic algorithm-enhanced artificial neural network-based CT radiomics signature can effectively distinguish the MSI status of CRC patients. • Kaplan-Meier survival analysis demonstrated that our signature could significantly stratify stage II CRC patients into high- and low-risk groups.

EPH109 A Scoping Review of Interactive Online Decision Tools to Guide Surgery and Adjuvant Treatment Decisions in Breast Cancer

To evaluate the interactive, currently available, web-based tools to support treatment and surgery decisions in clinical practice and inform the future directions in clinical tool development to support breast cancer care.

Trends in Pregnancy-Associated Cervical Cancer in Japan between 2012 and 2017: A Multicenter Survey.

Simple SummaryThe available evidence does not sufficiently indicate the status of pregnancy-associated cervical cancer in Japan and the associated treatment options. This study aimed to assess the occurrence of pregnancy-associated cervical cancer, available treatments, and the impact on the mother and the unborn child. The results show that pregnancy-associated cervical cancer is occurring more frequently in Japan, and most of the patients with early-stage disease have successful deliveries, although some have preterm births. Our findings also show that pregnant women with stage IB1 cervical cancer have the option of continuing pregnancy in addition to chemotherapy, as well as removal of the cervix and surrounding tissues. Thus, the treatment approach for pregnancy-associated cervical cancer must consider the stage of the cancer and the benefits that will accrue for both the mother and the unborn child.Large-scale data on maternal and neonatal outcomes of pregnancy-associated cervical cancer in Japan are scarce, and treatment strategies have not been established. This multicenter retrospective observational study investigated clinical features and trends in pregnancy-associated cervical cancer treatments at 523 hospitals in Japan. We included cervical cancer cases that were histologically diagnosed (between 1 January 2012, and 31 December 2017), and their clinical information was retrospectively collected. Of 40 patients diagnosed with pregnancy-associated cervical cancer at ≥22 gestational weeks, 34 (85.0%) were carefully followed until delivery without intervention. Of 163 diagnosed at <22 gestational weeks, 111 continued and 52 terminated their pregnancy. Ninety patients with stage IB1 disease had various treatment options, including termination of pregnancy. The 59 stage IB1 patients who continued their pregnancy were categorized by the primary treatment into strict follow-up, conization, trachelectomy, and neoadjuvant chemotherapy groups, with no significant differences in progression-free or overall survival. The birth weight percentile at delivery was smaller in the neoadjuvant chemotherapy group than in the strict follow-up group (p = 0.029). Full-term delivery rate was relatively higher in the trachelectomy group (35%) than in the other groups. Treatment decisions for pregnancy-associated cervical cancer are needed after estimating the stage, considering both maternal and fetal benefits.

Concordance of breast cancer biomarker testing in core needle biopsy and surgical specimens: A single institution experience.

Accurate diagnostic biomarker testing is crucial to treatment decisions in breast cancer. Biomarker testing is performed on core needle biopsies (CNB) and is often repeated in the surgical specimen (SS) after resection. As differences between CNB and SS testing may alter treatment decisions, we evaluated concordance between CNB and SS as well as associated changes in treatment and clinical outcomes. We performed a retrospective analysis of breast cancer patients at our institution between January 2010 and May 2020. Concordance between CNB and SS was assessed for estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) by immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH). Survival in patients, including recurrence, metastatic recurrence, and death, were assessed using chi-squared likelihood ratio. In total, 961 patients met eligibility criteria. Concordance, minor discordance, total concordance (concordance plus minor discordance), and major discordance between CNB and SS were reported for ER (87.7%, 9.2%, 90.8%, and 2.9%), PR (58.1%, 29.1%, 87.2%, and 12.8%), and HER2 IHC (52.5%, 20.9%, 73.4%, 26.6%), respectively. HER2 FISH concordance and major discordance were 58.5% and 1.2%, respectively. Of major discordance, ER (48.2%, p< 0.001) and HER2 FISH (50.0%) led to more management changes than HER2 IHC (2.4%, p= 0.04) and PR (1.6%, p= 0.10). Patients with ER major discordance had increased risk of death (6.7% concordance vs. 22.2% major discordance, p= 0.004). Overall, retesting ER and HER2 was more clinically beneficial than retesting PR. To aid decision-making and minimize healthcare costs, we propose patient-centered guidelines on retesting biomarker profiles.

Up-front cell-free DNA next generation sequencing improves target identification in UK first line advanced non-small cell lung cancer (NSCLC) patients

BackgroundGenomic sequencing is necessary for first-line advanced non-small cell lung cancer (aNSCLC) treatment decision-making. Tissue next generation sequencing (NGS) is standard but tissue quantity, quality, and time-to-results remains problematic. Here, we compare upfront cell-free-DNA (cfDNA) NGS clinical utility against routine tissue testing in patients with aNSCLC. MethodscfDNA-NGS was performed in consecutive, newly identified aNSCLC patients between December 2019–October 2021 alongside routine tissue genotyping. Variants were interpreted using AMP/ASCO/CAP guidelines. The primary endpoint was tier-1 variants detected on cfDNA-NGS. cfDNA-NGS results were compared to tissue results. ResultsOf 311 patients, 282 (91%) had an informative cfDNA-NGS test; 118 (38%) patients had a tier-1 variant identified by cfDNA-NGS. Of 243 patients with paired tissue-cfDNA tests, 122 (50%) tissue tests were informative; 85 (35%) tissue tests identified a tier-1 variant. cfDNA-NGS detected 39 additional tier-1 variants compared to tissue alone, increasing the tier-1 detection rate by 46% (from 85 to 124). The sensitivity of cfDNA-NGS relative to tissue was 75% (25% tissue tier-1 variants were not detected on cfDNA-NGS); 33% of cfDNA tier-1 variants were not identified on tissue tests. Median time from request-to-report was shorter for cfDNA-NGS versus tissue (8 versus 22 days; p < 0.0001).A total of 245 (79%) patients received first-line systemic-therapy: 49 (20%) with cfDNA-NGS results alone. Median time from sampling-to-commencement of first-line treatment was shorter for cfDNA-NGS blood draw versus first tissue biopsy (16 versus 35 days; p < 0.0001). ConclusionscfDNA-NGS increased the tier-1 variant detection rate with high concordance with tissue, and halves time-to-treatment. ‘Plasma-first’ upfront cfDNA-NGS use should be considered routinely for aNSCLC.

Does the addition of whole-body MRI to routine imaging influence real-world treatment decisions in metastatic breast cancer?

BackgroundThe assessment of metastatic breast cancer (MBC) can be limited with routine imaging such as computed tomography (CT) especially in bone-only or bone-predominant disease. This analysis investigates the effects of the use of WBMRI in addition to the use of routine CT, bone scintigraphy (BS) and fluorine-18-fluorodeoxyglucose positron emission tomography with computed tomography (FDG-PET/CT) on influencing systemic anti-cancer treatment (SACT) decisions in patients with known MBC.MethodsMBC patients undergoing SACT who had WBMRI undertaken within 8 weeks of either a routine CT, BS or FDG-PET/CT were reviewed retrospectively. The clinical indications for undertaking the WBMRI examinations were recorded. Data on the extent and distribution of the disease were collected and discordance/concordance of disease status across the imaging modalities were compared. SACT decisions at each time point were also evaluated.ResultsThere were 105 MBC patients with 148 WBMRI studies paired with CT, BS or FDG-PET/CT. 50 pairs (33.8%) showed differences in the extent of disease, with 44 pairs due to additional sites (AS) reported on WBMRI alone.81 patients (Group 1) had one WBMRI paired with routine imaging due to a variety of indications, with clinical symptoms (such as bone pain) being the most common (24.7%).24 patients (Group 2) had more than one WBMRI study paired with routine imaging comprising 67 pairs. 13/67 pairs (19.4%) showed discordance in assessments. 10/13 pairs had progressive disease (PD) reported on WBMRI alone.SACT change due to AS reported on WBMRI alone occurred in 21/23 pairs (91.3%) in Group 1. SACT change due to PD reported on WBMRI alone in Group 2 occurred in 6/14 pairs (42.9%). SACT change due to AS/PD in both groups occurred in 11/102 pairs (10.8%) with known invasive ductal carcinoma (IDC) and 13/28 pairs (46.4%) with invasive lobular carcinoma (ILC).ConclusionsThe use of WBMRI in MBC led to earlier recognition of PD and SACT change compared with the other imaging modalities. A higher proportion of discordant response assessments and SACT changes were observed in ILC compared with IDC in our patient group, although larger-scale studies are required to investigate this further.

Validation of the prostate cancer comorbidity index in predicting cause-specific mortality in men undergoing radical prostatectomy.

Accurate prediction of competing risks of mortality remains a key component of prostate cancer treatment decision-making. We sought to validate the Prostate Cancer Comorbidity Index (PCCI) score for predicting other-cause mortality (OCM) and cancer outcomes in men undergoing radical prostatectomy (RP). We sampled 4857 men with prostate cancer treated with RP in the VA from 2000-2018. Risks of OCM, 90-day all-cause mortality (ACM), prostate cancer-specific mortality, metastasis, and biochemical recurrence by PCCI score were assessed using Cox proportional hazards and logistic regression. We compared prediction of 90-day ACM between PCCI and the American Society of Anesthesiology (ASA) score, a validated predictor of short-term mortality. Over median follow-up of 6.7 years (IQR 3.7-10.3), there was a stepwise increase in risk of OCM with higher PCCI score, with hazards (95%CI) of 1.53 (1.14-2.04), 2.11 (1.55-2.88), 2.36 (1.68-3.31), 3.61 (2.61-4.98), and 4.99 (3.58-6.96) for PCCI 1-2, 3-4, 5-6, 7-9, and 10 + (vs. 0), respectively. Projected 10-year cumulative incidence of OCM was 8%, 12%, 16%, 19%, 26%, and 32% for scores of 0, 1-2, 3-4, 5-6, 7-9, and 10+ , respectively. Men with PCCI 7+ had greater odds of 90-day ACM (OR 3.48, 95%CI 1.26-9.63) while men with higher ASA did not. Higher PCCI score was associated with worse cancer outcomes, with the highest categories driving the associations. The PCCI is a robust measure of short- and long-term OCM after RP, validated for use in clinical care and health services research focusing on surgical patient populations.

P-200 Mutational testing on liquid biopsies for treatment decisions in metastatic colorectal cancer – comparison of ddPCR and MassARRAY methods

The aim was to investigate if mutational analysis of circulating DNA (cfDNA) in plasma is feasible for selection of patients with metastatic colorectal cancer (mCRC) for anti-EGFR treatment. Secondary, to investigate emergence of mutations in cfDNA during therapy, and to compare the fractions of mutations detected by the ddPCR and MassARRAY methods. This was a prospective clinical feasibility study using circulating tumor DNA (ctDNA) for treatment selection in mCRC. Inclusion criteria: mCRC; indication for systemic palliative treatment; measureable disease (RECIST v1.1); age ≥ 18. Blood samples were drawn at inclusion, during treatment and follow up. Treatment regimens according to standard. Samples for prospective analyzes were collected in Streck tubes, shipped to central laboratory and DNA analyzed by a multiplex ddPCR (Bio-Rad) testing for 30 somatic mutations in KRAS/NRAS/BRAF with a limit of detection of app. 0.01% as previously published. Feasibility endpoints were; quality and turn-around time of ctDNA analysis. Failure parameters were; low quality of samples (lymphocyte contamination > 5%, major loss 3 working days (WD), results delivered > 7 WD. For retrospective analyzes plasma samples were collected in EDTA tubes and analyzed by a mass spectrometric based multiplexed platform (MassARRAY® Agena Bioscience). The UltraSEEK MassARRAY Colon Panel v2 was used to test for 104 somatic mutations in KRAS/NRAS/BRAF/EGFR/PIK3CA with a limit of detection of app. 0.01%. All mutations targeted by the ddPCR panel were included. Mutational testing on tumor tissue was done according to standard. Forty-nine patients were included. Average time from sampling to results by ddPCR were 2.7 WD. One sample failed quality parameters due to transportation to lab > 3 WD. Concordance between mutational status in tissue and plasma by ddPCR was 70%. Disconcordant (30%) were wild type (wt) in plasma, but mutated in tumor tissue. Comparing the concordant and disconcordant cases there was no difference in treatment response (p=0.46) but longer progression free survival (HR=2.1 95% CI 1.05-4.27, p=0.04) and a tendency to longer overall survival (HR=2.3 95% CI 0.97-5.32, p=0.06) among disconcordant cases. Concordance between mutational status in tissue and plasma evaluated by MassARRAY was 80%. When comparing mutational testing by ddPCR and MassARRAY at inclusion, all but one mutation detected by ddPCR were detected by MassARRAY. In four patients, who were wt by ddPCR, MassARRAY detected a mutation, which were in the ddPCR panel and also present in the tumor tissue. MassARRAY detected mutations in seven patients, which were not covered by the ddPCR panel. In one patient we observed the emergence of a single mutation (EGFR S492R, by MassARRAY) at the end-of-study. Using cfDNA for mutational detection by ddPCR in daily clinic proved feasible. Treatment response and survival among disconcordant cases were not inferior to concordant cases. Mutational testing by MassARRAY seemed to have a better concordance with tissue status than ddPCR and future studies should investigate MassARRAY for selection of patients with mCRC who will benefit from anti-EGFR treatment. Emergence of a new mutation during systemic palliative treatment was a rare event.

Frailty status derived from the electronic medical record to predict survival and health care utilization in veterans with lung and gastrointestinal cancers.

6582 Background: Frailty is associated with increased vulnerability to stressors. Although frailty is a predictor of survival, traditional frailty assessments are not feasible in clinical practice. Frailty indexes (FI) derived from the electronic medical record (EMR) are an alternate approach. There is limited data for the association between EMR-FI and cancer-related outcomes including survival and health care utilization. Methods: We identified a cohort of patients diagnosed with lung or gastrointestinal cancer from 2016-2020 using VA administrative files. We used the Veterans Affairs-FI (VA-FI), a validated 31-item cumulative deficit FI, to define three groups: robust (≤0.1), prefrail (0.1-0.2), frail ( &gt; 0.2). Cox proportional hazard analyses were conducted to evaluate survival. Logistic regression analyses were performed to examine healthcare utilization. All models were adjusted for age, gender, race, Charlson comorbidity index, and stage. Results: Among 37,439 patients (age: 70±8.6 years), the majority were prefrail (36.8%) or frail (24.1%). Prefrail patients had a 13% increased risk of death compared to robust patients (adjusted hazard ratio [aHR] 1.13; 95% CI 1.10-1.16), while frail patients had a 52% increased risk (aHR 1.52; 95% CI 1.47, 1.57). Pre-frail and frail patients had increased odds of 1-year ED visits (adjusted odds ratio [aOR] 1.33; 95% CI 1.27 - 1.40 and aOR, 1.88, 95% CI 1.77-2.00, respectively). Frailty status was associated with increased odds of 1-year all-cause hospitalization among prefrail (aOR 1.09; 95% CI 1.03-1.15) and frail patients (aOR 1.26; 95%CI: 1.18, 1.34). Conclusions: Frailty measured by EMR data is significantly associated with survival and healthcare utilization among patients with lung and gastrointestinal cancers, independent of stage. Further study is warranted to develop EMR-FI as a risk stratification tool for personalized cancer treatment decisions.

Expanding the precision oncology toolkit with micro-organospheres for early cancer diagnosis.

Using an automatic microfluidics droplet platform, Ding etal. successfully replicated the tumor micro-environment by generating micro-organospheres, which were then used to predict the response to anti-tumor drugs. These miniature models could be obtained within an extremely short time frame of 14days, amplifying their role in facilitating cancer treatment decisions.

Influence of decision support persons on breast cancer treatment decisions among Latinas.

12025 Background: Prior studies suggest that decision support persons (DSPs) involvement in breast cancer promotes greater deliberation and decision quality. Despite having the highest level of involvement, Latinx DSPs report the lowest satisfaction with their involvement. The reasons for this remain unknown. We examined the treatment decision-making experiences of Latinx DSPs, their influence on treatment deliberation, subjective decision quality (SDQ), and treatment received. Methods: Women with newly diagnosed early-stage breast cancer as reported to the Surveillance, Epidemiology, and End Results (SEER) registries of Georgia and Los Angeles County in 2014-2015 were surveyed. Participants identified the DSPs who played a key role in treatment decisions, who were also surveyed. We examined: (1) bivariate associations of DSP characteristics (e.g type, age, race/ethnicity, education, acculturation level) with DSP-reported level of engagement (informed about decisions, involvement (extent and satisfaction), and aware of patient preferences), (2) DSP engagement with patient-reported SDQ and treatment deliberation using multivariable linear regression with standardized scales (3), and treatment received by DSPs preferred treatment. Results: 2502 patients (68%) and 1203 eligible DSPs (70%) responded, resulting in 1,173 dyads, 292 where the patient identified as Latina, and 881 as non-Latina. Among Latina dyads, 78%, 17%, and 5% DSPs identified as Latinx, White, and Asian/Black/Other, respectively. Latinx DSPs within Latina/Latinx dyads were younger, had lower educational attainment and acculturation when compared to other dyads. The proportion of married/partnered status was not different across dyads, but the key DSP for the Latina-Latinx dyads was more often a daughter (37%), over a husband/partner (21%), compared to the other dyads. Latinx DSPs reported being more informed (adjusted mean 4.26, p = 0.058) compared to the other dyads, and being more informed was positively associated with higher patient SDQ (adjusted mean difference 0.176, p = 0.034), despite no difference in treatment deliberation. Overall, Latinx DSP had a higher preference for mastectomy, especially with reconstruction when compared to non-Latinx (40% vs 28%). Overall, 27% of Latinas (vs 13% non-Latina) underwent lumpectomy despite their DSP’s preference for mastectomy. Conclusions: These findings reveal that the key DSP for many married Latinas is often a daughter over spouse/partner. Our results suggest that including daughter/DSPs in treatment decisions and tailoring strategies to meet their information needs may positively impact Latina SDQ. Potential areas of improvement include surgical options preferences, where notable discrepancy was seen between Latinas and non-Latinas. Awareness of these differences can minimize treatment regret, improve decision quality, and ultimately outcomes in Latinas.

“Like Not Having an Arm”: Perspectives of Patients, Caregivers, and Practitioners on the Impact of Visitor Restrictions on Cancer Care During the COVID-19 Pandemic (RP405)

Outcomes1. Understand how caregiver absence affected the patient and caregiver dyadic relationship2. Understand how caregiver absence affected cancer and palliative care practitionersImportanceVisitor restriction policies to prevent the spread of COVID-19 between patients and practitioners have been widespread during the pandemic, resulting in the exclusion of caregivers at key points of cancer care and treatment decision making.Objective(s)Explore how visitor restrictions impacted cancer treatment decision making and care from patient, caregiver, and practitioner perspectives.Method(s)71 interviews, including 48 cancer patients, 4 family caregivers, and 19 cancer and palliative care practitioners from four academic cancer centers in the US between August 2020 and July 2021.ResultsVisitor restrictions that prevented caregivers from participating in clinic appointments and perioperative hospital stays were universally disliked by patients, caregivers, and practitioners. Patients felt deprived of an advocate to help ask questions and to process and retain medical information when patients felt overwhelmed or impaired with “chemo brain.” Caregivers discussed that visitor restrictions made them feel “out of the loop” and increased their anxiety about the patient's care. Cancer practitioners described how visitor restrictions impaired care because caregivers were not present to share valuable health information that the patient may have forgotten or neglected to mention, and clinicians had to spend extra time to call caregivers to relay information. Some surgical oncologists expressed moral distress because they perceived that the absence of caregivers negatively affected their patients’ health outcomes because caregivers were not there to provide emotional support during hospital stays.Conclusion(s)Our data indicate that the structure of the medical decision-making relationship is more tripartite than bilateral, with caregivers being a significant third actor within the medical decision-making process. Caregivers provided support in the decision-making and care process to both patients and practitioners, and their absence created additional burden on both.ImpactAs the pandemic continues into its second year with new surges and a return to visitor restrictions, health systems should reconsider visitor policies to ease the emotional and physical burden on patients, caregivers, and practitioners.

Relative Incidence of Emergency Department Visits After Treatment for Prostate Cancer With Radiation Therapy or Radical Prostatectomy

PurposeSide effect profiles play an important role in treatment decisions for localized prostate cancer. Emergency department (ED) visits, which may be due to side effects from treatment, can be measured in real-world, structured, electronic health record (EHR) data. The goal of this study was to determine whether treatments for localized prostate cancer are associated with ED visits, as a measure of side effects, using EHR data. Methods and MaterialsWe used a self-controlled case series study design, including patients treated at an urban academic medical center with radiation therapy (RT) or radical prostatectomy (RP) for prostate cancer between 2011 and 2020 who had visits documented for ≥6 months before and after treatment and ≥1 ED visit. We estimated relative incidences (RI) of ED visits, comparing incidence in the exposed and unexposed periods, with the exposed period being between start of treatment and 1 month after completion, and the unexposed period consisting of all other documented time. ResultsAmong men who had at least 1 ED visit and after adjusting for age, there were higher rates of ED visits after RP (RI, 20.4; 95% confidence interval [CI], 15.4-27.0; P < .001), RT overall (RI, 2.4; CI, 1.7-3.4; P < .001), intensity modulated radiation therapy with high dose-rate brachytherapy (RI, 3.4; CI, 1.7-6.8; P < .001) or stereotactic body radiation therapy boost (RI, 7.1; CI, 3.4-14.8; P < .001), and high dose rate brachytherapy alone (RI, 16.3; CI, 7.2-36.9; P < .001) compared with unexposed time. The number needed to harm to result in an ED visit was less for RP (17; CI, 13-23) than RT overall (43; CI, 25-126), but varied by RT modality. ConclusionsIn summary, relative rates of ED visits vary by treatment type, suggesting differing severities of side effects. These data may aid in selecting treatments and demonstrate the feasibility of using the self-controlled case series study design on ED visits in real-world, structured EHR data to better understand side effects of treatment.

Hubness weighted SVM ensemble for prediction of breast cancer subtypes

BACKGROUND: Breast cancer is a major disease causing panic among women worldwide. Since gene mutations are the root cause for cancer development, analyzing gene expressions can give more insights into various phenotype of cancer treatments. Breast Cancer subtype prediction from gene expression data can provide more information for cancer treatment decisions. OBJECTIVE: Gene expressions are complex for analysis due to its high dimensional nature. Machine learning algorithms such as k-Nearest Neighbors, Support Vector Machine (SVM) and Random Forest are used with selection of features for prediction of breast cancer subtypes. Prediction accuracy of the existing methods are affected due to high dimensional nature of gene expressions. The objective of the work is to propose an efficient algorithm for the prediction of breast cancer subtypes from gene expression. METHODS: For subtype prediction, a novel Hubness Weighted Support Vector machine algorithm (HWSVM) using bad hubness score as a weight measure to handle the outliers in the data has been proposed. Based on the various subtypes, features are projected into seven different feature sets and Ensemble based Hubness Aware Weighted Support Vector Machine (HWSVMEns) is implemented for breast cancer subtype prediction. RESULTS: The proposed algorithms have been compared with the classical SVM and other traditional algorithms such as Random Forest, k-Nearest Neighbor algorithms and also with various gene selection methods. CONCLUSIONS: Experimental results show that the proposed HWSVM outperforms other algorithms in terms of accuracy, precision, recall and F1 score due to the hubness weightage scheme and the ensemble approach. The experiments have shown an average accuracy of 92% across various gene expression datasets.

Clinician perspectives on clinical decision support systems in lung cancer: Implications for shared decision-making.

BackgroundLung cancer treatment decisions are typically made among clinical experts in a multidisciplinary tumour board (MTB) based on clinical data and guidelines. The rise of artificial intelligence and cultural shifts towards patient autonomy are changing the nature of clinical decision‐making towards personalized treatments. This can be supported by clinical decision support systems (CDSSs) that generate personalized treatment information as a basis for shared decision‐making (SDM). Little is known about lung cancer patients' treatment decisions and the potential for SDM supported by CDSSs. The aim of this study is to understand to what extent SDM is done in current practice and what clinicians need to improve it.ObjectiveTo explore (1) the extent to which patient preferences are taken into consideration in non‐small‐cell lung cancer (NSCLC) treatment decisions; (2) clinician perspectives on using CDSSs to support SDM.DesignMixed methods study consisting of a retrospective cohort study on patient deviation from MTB advice and reasons for deviation, qualitative interviews with lung cancer specialists and observations of MTB discussions and patient consultations.Setting and ParticipantsNSCLC patients (N = 257) treated at a single radiotherapy clinic and nine lung cancer specialists from six Dutch clinics.ResultsWe found a 10.9% (n = 28) deviation rate from MTB advice; 50% (n = 14) were due to patient preference, of which 85.7% (n = 12) chose a less intensive treatment than MTB advice. Current MTB recommendations are based on clinician experience, guidelines and patients' performance status. Most specialists (n = 7) were receptive towards CDSSs but cited barriers, such as lack of trust, lack of validation studies and time. CDSSs were considered valuable during MTB discussions rather than in consultations.ConclusionLung cancer decisions are heavily influenced by clinical guidelines and experience, yet many patients prefer less intensive treatments. CDSSs can support SDM by presenting the harms and benefits of different treatment options rather than giving single treatment advice. External validation of CDSSs should be prioritized.Patient or Public ContributionThis study did not involve patients or the public explicitly; however, the study design was informed by prior interviews with volunteers of a cancer patient advocacy group. The study objectives and data collection were supported by Dutch health care insurer CZ for a project titled ‘My Best Treatment’ that improves patient‐centeredness and the lung cancer patient pathway in the Netherlands.

The benefits of G8 and VES-13 geriatric screening tools for older patients with advanced lung cancer

Cancer is a disease of ageing. There is scarce evidence to guide proper cancer treatment decisions for this population because older patients, especially those with age-related comorbidities, are under-represented in clinical trials [1]. During the past decades, there have been major breakthroughs in non-small cell lung cancer treatment. Up to half of the Asian patients with adenocarcinoma harbor epidermal growth factor receptor (EGFR) mutations – the predictive biomarker of EGFR tyrosine kinase inhibitors (EGFR TKIs) [2].

Communication, perception, and use of personalized side-effect risks in prostate cancer treatment-decision making: An observational and interview study

ObjectiveWe investigated how healthcare professionals (HPs) communicate personalized risks of treatment side-effects to patients with localized prostate cancer during consultations, and explored how these patients perceive and use such risks during treatment decision-making. MethodsPatient consultations with nurse practitioners and urologists discussing personalized risks of urinary incontinence after prostatectomy were audiotaped, transcribed, and coded. Patients (n = 27) were then interviewed to explore their perceptions and use of personalized side-effect risks. ResultsHPs explained personalized risks by discussing risk factors, which was appreciated and recalled by patients. Personalized risks were typically communicated both numerically and verbally (70%). When using numbers, HPs always used percentages, but rarely used natural frequencies (14%). Uncertainty was disclosed in only 34% of consultations. One-third of patients used personalized risks in their treatment decision-making by either switching to another treatment or sticking to their initial preference. ConclusionsPatients value and use personalized side-effect risks during treatment decision-making. Clearly explaining the relationship between risk factors and personalized risk estimates may help patients understand and recall those.Practice implicationsHPs should not only give patients specific and precise numerical risk information, but should also put effort in explaining how the personalized side-effect risks are determined.

Time trends in cancer and dementia related hospital admissions among Medicare fee-for-service population, 2013–2018

Cancer and dementia are two complicated morbidities that are commonly found in older adults and require substantial healthcare resources.[ [1] Caba Y. et al. The impact of dementia on cancer treatment decision-making, cancer treatment, and mortality: a mixed studies review. JNCI Cancer Spectr. 2021; 5: 1-18 Crossref Google Scholar , [2] McWilliams L. et al. A systematic review of the prevalence of comorbid cancer and dementia and its implications for cancer-related care. Aging Ment Health. 2017; : 1-18 Google Scholar ] Caring for patients with the co-presence of cancer and dementia, who suffer from impaired cognitive function with complex medical needs, is challenging for clinicians, caregivers, and healthcare systems. With the aging US population, the increasing number of older adults with these two comorbid conditions is likely to place a high demand on hospitals. Meanwhile, the shift of care delivery from inpatient to settings such as outpatient and homes, increasing reliance on family caregivers, and improved community care support may help reduce unnecessary hospitalizations.[ 3 2021 Alzheimer's disease facts and figures Alzheimers Dement. 2021; 17: 327-406 Crossref PubMed Scopus (739) Google Scholar , 4 Federman A.D. et al. Association of a bundled hospital-at-home and 30-day postacute transitional care program with clinical outcomes and patient experiences. JAMA Intern Med. 2018; 178: 1033-1040 Crossref PubMed Scopus (88) Google Scholar , 5 Handley N.R. Bekelman J.E. The oncology hospital at home. J Clin Oncol. 2019; 37: 448-452 Crossref PubMed Scopus (23) Google Scholar ] It is important to understand the pattern of hospitalization, which is one of the major drivers of healthcare costs and high financial and caregiving burden for patients and their families, among the growing population with cancer and dementia. While there has been extensive research on the incidence, prevalence, and healthcare utilization among older adults with either cancer or dementia, existing literature examining older adults with both conditions is scarce and limited in scope. [ [1] Caba Y. et al. The impact of dementia on cancer treatment decision-making, cancer treatment, and mortality: a mixed studies review. JNCI Cancer Spectr. 2021; 5: 1-18 Crossref Google Scholar , [2] McWilliams L. et al. A systematic review of the prevalence of comorbid cancer and dementia and its implications for cancer-related care. Aging Ment Health. 2017; : 1-18 Google Scholar , [6] Kedia S.K. et al. Health care utilization among elderly medicare beneficiaries with coexisting dementia and cancer. Gerontol Geriatr Med. 2017; 3 (p. 2333721416689042) Crossref PubMed Google Scholar , [7] Liu B. et al. Trends of hospitalizations among patients with both cancer and dementia diagnoses in New York 2007-2017. Healthc (Amst). 2021; 9100565 Google Scholar ] In particular, no national estimates exist regarding the hospitalization rates among older adults with comorbid cancer and dementia. To fill this gap, we examined the national and state-level trends in hospitalization rates among Medicare fee-for-service (FFS) beneficiaries with the co-presence of a major cancer (breast, prostate, lung, and colorectal) and dementia diagnoses. We hypothesized a growing trend of inpatient use among this population during the years 2013–2018. Due to the heterogeneity in cancer-specific prognosis and treatment, we also examined the trend of hospitalizations among this population by cancer types.

70P Factors influencing patient treatment decisions in early breast cancer (eBC): Discrete choice experiment (DCE) findings

The emerging treatment paradigm in eBC, with new systemic therapies varying in efficacy, safety and conditions of use, creates a complex patient journey involving many inflection points with respect to treatment sequence, duration and response to initial therapy. We quantified patient preferences for attributes of different treatment pathways in eBC in Germany, Italy and Japan.