JCO Clinical Cancer Informatics

Altrock, Philipp M.; Ferlic, Jeremy; Galla, Tobias; Tomasson, Michael H.; Michor, Franziska

2018 JCO Clinical Cancer Informatics

doi: 10.1200/CCI.17.00131pmid: 30652561

Purpose: Recent advances have uncovered therapeutic interventions that might reduce the risk of progression of premalignant diagnoses, such as monoclonal gammopathy of undetermined significance (MGUS) to multiple myeloma (MM). It remains unclear how to best screen populations at risk and how to evaluate the ability of these interventions to reduce disease prevalence and mortality at the population level. To address these questions, we developed a computational modeling framework. Materials and Methods: We used individual-based computational modeling of MGUS incidence and progression across a population of diverse individuals to determine best screening strategies in terms of screening start, intervals, and risk-group specificity. Inputs were life tables, MGUS incidence, and baseline MM survival. We measured MM-specific mortality and MM prevalence after MGUS detection from simulations and mathematic modeling predictions. Results: Our framework is applicable to a wide spectrum of screening and intervention scenarios, including variation of the baseline MGUS to MM progression rate and evolving MGUS, in which progression increases over time. Given the currently available point estimate of progression risk reduction to 61% risk, starting screening at age 55 years and performing follow-up screening every 6 years reduced total MM prevalence by 19%. The same reduction could be achieved with starting screening at age 65 years and performing follow-up screening every 2 years. A 40% progression risk reduction per patient with MGUS per year would reduce MM-specific mortality by 40%. Specifically, screening onset age and screening frequency can change disease prevalence, and progression risk reduction changes both prevalence and disease-specific mortality. Screening would generally be favorable in high-risk individuals. Conclusion: Screening efforts should focus on specifically identified groups with high lifetime risk of MGUS, for which screening benefits can be significant. Screening low-risk individuals with MGUS would require improved preventions.

Henry, Lynn N.; Kidwell, Kelley M.; Alsamarraie, Cindy; Bridges, Celia M.; Kwiatkowski, Christine; Clauw, Daniel J.; Smith, Ellen M. L.; Williams, David A.

2018 JCO Clinical Cancer Informatics

doi: 10.1200/CCI.17.00106pmid: 30652562

Purpose: Many survivors of breast cancer experience an array of chronic symptoms, including pain, insomnia, and fatigue. Few effective therapies have been identified. Behavioral management programs to address similar symptom clusters in other chronic conditions have been effective. The objective of this study was to determine the effect of an Internet-based lifestyle and behavioral self-management program on cancer-related symptoms. Patients and Methods: Women with stage 0 to 3 breast cancer who reported insomnia, pain, or fatigue as their primary symptom of concern during the 7 days before enrollment were enrolled. Local therapies and/or chemotherapy were completed at least 3 months before enrollment. Patients were assessed at baseline and after 8 weeks, and they completed the Patient-Reported Outcomes Measurement Information System (PROMIS)-29 Profile and Patient Global Impression of Change (PGIC) questionnaire electronically. Change in each of the eight symptom domains was assessed. Results: Fifty patients enrolled. In the 45 patients with both baseline and 8-week PROMIS data, statistically significant improvements in anxiety, sleep, fatigue, activity level, and pain severity were reported. Of the 35 patients who responded to the PGIC, 62.9% reported improvement in their primary symptom. Those who reported fatigue as their primary symptom reported greatest overall benefit in multiple symptom improvement, including improvements in fatigue, anxiety, pain severity, pain interference, and participation in social activities. Conclusions: These findings suggest that this lifestyle and behavioral management program may improve multiple symptoms in breast cancer survivors when delivered via the Internet. Randomized studies are warranted to evaluate the efficacy of the online intervention compared with standard symptom management approaches and to identify patients most likely to benefit.

Renshaw, Andrew A.; Mena-Allauca, Mercy; Gould, Edwin W.; Sirintrapun, Joseph S.

2018 JCO Clinical Cancer Informatics

doi: 10.1200/CCI.17.00088pmid: 30652566

Krauss, John C.; Sahai, Vaibhav; Kirch, Matthias; Simeone, Diane M.; An, Lawrence

2018 JCO Clinical Cancer Informatics

doi: 10.1200/CCI.17.00086pmid: 30652554

Purpose: The treatment of cancer is complex, which can overwhelm patients and lead to poor comprehension and recall of the specifics of the cancer stage, prognosis, and treatment plan. We hypothesized that an oncologist can feasibly record and deliver a custom video summary of the consultation that covers the diagnosis, recommended testing, treatment plan, and follow-up in < 5 minutes. The video summary allows the patient to review and share the most important part of a cancer consultation with family and caregivers. Methods: At the conclusion of the office visit, oncologists recorded the most important points of the consultation, including the diagnosis and management plan as a short video summary. Patients were then e-mailed a link to a secure Website to view and share the video. Patients and invited guests were asked to respond to an optional survey of 15 multiple-choice and four open-ended questions after viewing the video online. Results: Three physicians recorded and sent 58 video visit summaries to patients seen in multidisciplinary GI cancer clinics. Forty-one patients logged into the secure site, and 38 viewed their video. Fourteen patients shared their video and invited a total of 46 visitors, of whom 36 viewed the videos. Twenty-six patients completed the survey, with an average overall video satisfaction score of 9 on a scale of 1 to 10, with 10 being most positive. Conclusion: Video visit summaries provide a personalized education tool that patients and caregivers find highly useful while navigating complex cancer care. We are exploring the incorporation of video visit summaries into the electronic medical record to enhance patient and caregiver understanding of their specific disease and treatment.

Rayfield, Corbin A.; Grady, Fillan; De Leon, Gustavo; Rockne, Russell; Carrasco, Eduardo; Jackson, Pamela; Vora, Mayur; Johnston, Sandra K.; Hawkins-Daarud, Andrea; Clark-Swanson, Kamala R.; Whitmire, Scott; Gamez, Mauricio E.; Porter, Alyx; Hu, Leland; Gonzalez-Cuyar, Luis; Bendok, Bernard; Vora, Sujay; Swanson, Kristin R.

2018 JCO Clinical Cancer Informatics

doi: 10.1200/CCI.17.00080pmid: 30652553

Purpose: Despite the intra- and intertumoral heterogeneity seen in glioblastoma multiforme (GBM), there is little definitive data on the underlying cause of the differences in patient survivals. Serial imaging assessment of tumor growth allows quantification of tumor growth kinetics (TGK) measured in terms of changes in the velocity of radial expansion seen on imaging. Because a systematic study of this entire TGK phenotype-growth before treatment and during each treatment to recurrence -has never been coordinately studied in GBMs, we sought to identify whether patients cluster into discrete groups on the basis of their TGK. Patients and Methods: From our multi-institutional database, we identified 48 patients who underwent maximally safe resection followed by radiotherapy with imaging follow-up through the time of recurrence. The patients were then clustered into two groups through a k-means algorithm taking as input only the TGK before and during treatment. Results: There was a significant survival difference between the clusters (P = .003). Paradoxically, patients among the long-lived cluster had significantly larger tumors at diagnosis (P = .027) and faster growth before treatment (P = .003) but demonstrated a better response to adjuvant chemotherapy (P = .048). A predictive model was built to identify which cluster patients would likely fall into on the basis of information that would be available to clinicians immediately after radiotherapy (accuracy, 90.3%). Conclusion: Dichotomizing the heterogeneity of GBMs into two populations-ne faster growing yet more responsive with increased survival and one slower growing yet less responsive with shorter survival-suggests that many patients who receive standard-of-care treatments may get better benefit from select alternative treatments.

Lee, Jeeyun; An, Ji Yeong; Choi, Min Gew; Park, Se Hoon; Kim, Seung Tae; Lee, Jun Ho; Sohn, Tae Sung; Bae, Jae Moon; Kim, Sung; Lee, Hyuk; Min, Byung-Hoon; Kim, Jae J.; Jeong, Woo Kyoung; Choi, Dong-Il; Kim, Kyoung-Mee; Kang, Won Ki; Kim, Mijung; Seo, Sung Wook

2018 JCO Clinical Cancer Informatics

doi: 10.1200/CCI.17.00065pmid: 30652558

Purpose: Gastric cancer (GC) is the third-leading cause of cancer-related deaths. Several pivotal clinical trials of adjuvant treatments were performed during the previous decade; however, the optimal regimen for adjuvant treatment of GC remains controversial. Patients and Methods: We developed a novel deep learning-based survival model (survival recurrent network [SRN]) in patients with GC by including all available clinical and pathologic data and treatment regimens. This model uses time-sequential data only in the training step, and upon being trained, it receives the initial data from the first visit and then sequentially predicts the outcome at each time point until it reaches 5 years. In total, 1,190 patients from three cohorts (the Asian Cancer Research Group cohort, n = 300; the fluorouracil, leucovorin, and radiotherapy cohort, n = 432; and the Adjuvant Chemoradiation Therapy in Stomach Cancer cohort, n = 458) were included in the analysis. In addition, we added Asian Cancer Research Group molecular classifications into the prediction model. SRN simulated the sequential learning process of clinicians in the outpatient clinic using a recurrent neural network and time-sequential outcome data. Results: The mean area under the receiver operating characteristics curve was 0.92 +/- 0.049 at the fifth year. The SRN demonstrated that GC with a mesenchymal subtype should elicit a more risk-adapted postoperative treatment strategy as a result of its high recurrence rate. In addition, the SRN found that GCs with microsatellite instability and GCs of the papillary type exhibited significantly more favorable survival outcomes after capecitabine plus cisplatin chemotherapy alone. Conclusion: Our SRN predicted survival at a high rate, reaching 92% at postoperative year 5. Our findings suggest that SRN-based clinical trials or risk-adapted adjuvant trials could be considered for patients with GC to investigate more individualized adjuvant treatments after curative gastrectomy.

Dercle, Laurent; Connors, Dana E.; Tang, Ying; Adam, Stacey J.; Gonen, Mithat; Hilden, Patrick; Karovic, Sanja; Maitland, Michael; Moskowitz, Chaya S.; Kelloff, Gary; Zhao, Binsheng; Oxnard, Geoffrey R.; Schwartz, Lawrence H.

2018 JCO Clinical Cancer Informatics

doi: 10.1200/CCI.17.00137pmid: 30652552

Purpose: To develop a public-private partnership to study the feasibility of a new approach in collecting and analyzing clinically annotated imaging data from landmark phase III trials in advanced solid tumors. Patients and Methods: The collection of clinical trials fulfilled the following inclusion criteria: completed randomized trials of > 300 patients, highly measurable solid tumors (non-small-cell lung cancer, colorectal cancer, renal cell cancer, and melanoma), and required sponsor and institutional review board sign-offs. The new approach in analyzing computed tomography scans was to transfer to an academic image analysis laboratory, draw contours semi-automatically by using in-house-developed algorithms integrated into the open source imaging platform Weasis, and perform serial volumetric measurement. Results: The median duration of contracting with five sponsors was 12 months. Ten trials in 7,085 patients that covered 12 treatment regimens across 20 trial arms were collected. To date, four trials in 3,954 patients were analyzed. Source imaging data were transferred to the academic core from 97% of trial patients (n = 3,837). Tumor imaging measurements were extracted from 82% of transferred computed tomography scans (n = 3,162). Causes of extraction failure were nonmeasurable disease (n = 392), single imaging time point (n = 224), and secondary captured images (n = 59). Overall, clinically annotated imaging data were extracted in 79% of patients (n = 3,055), and the primary trial end point analysis in each trial remained representative of each original trial end point. Conclusion: The sharing and analysis of source imaging data from large randomized trials is feasible and offer a rich and reusable, but largely untapped, resource for future research on novel trial-level response and progression imaging metrics.

Zomnir, Michael G.; Lipkin, Lev; Pacula, Maciej; Dominguez Meneses, Enrique; MacLeay, Allison; Duraisamy, Sekhar; Nadhamuni, Nishchal; Al Turki, Saeed H.; Zheng, Zongli; Rivera, Miguel; Nardi, Valentina; Dias-Santagata, Dora; Iafrate, John A.; Le, Long P.; Lennerz, Jochen K.

2018 JCO Clinical Cancer Informatics

doi: 10.1200/CCI.16.00079pmid: 30364844

Purpose: Next-generation sequencing technologies are actively applied in clinical oncology. Bioinformatics pipeline analysis is an integral part of this process; however, humans cannot yet realize the full potential of the highly complex pipeline output. As a result, the decision to include a variant in the final report during routine clinical sign-out remains challenging. Methods: We used an artificial intelligence approach to capture the collective clinical sign-out experience of six board-certified molecular pathologists to build and validate a decision support tool for variant reporting. We extracted all reviewed and reported variants from our clinical database and tested several machine learning models. We used 10-fold cross-validation for our variant call prediction model, which derives a contiguous prediction score from 0 to 1 (no to yes) for clinical reporting. Results: For each of the 19,594 initial training variants, our pipeline generates approximately 500 features, which results in a matrix of > 9 million data points. From a comparison of naive Bayes, decision trees, random forests, and logistic regression models, we selected models that allow human interpretability of the prediction score. The logistic regression model demonstrated 1% false negativity and 2% false positivity. The final models' Youden indices were 0.87 and 0.77 for screening and confirmatory cutoffs, respectively. Retraining on a new assay and performance assessment in 16,123 independent variants validated our approach (Youden index, 0.93). We also derived individual pathologist-centric models (virtual consensus conference function), and a visual drill-down functionality allows assessment of how underlying features contributed to a particular score or decision branch for clinical implementation. Conclusion: Our decision support tool for variant reporting is a practically relevant artificial intelligence approach to harness the next-generation sequencing bioinformatics pipeline output when the complexity of data interpretation exceeds human capabilities.

Chen, Allen M.; Kupelian, Patrick A.; Wang, Pin-Cheih; Steinberg, Michael L.

2018 JCO Clinical Cancer Informatics

doi: 10.1200/CCI.17.00036pmid: 30652556

Purpose: The computerized paperless medical recording system has transformed the modern health information system and serves as an idea platform for registry development, particularly in a specialty such as radiation oncology, where technologic advances continue to generate unprecedented amounts of complex data. We present our single-institution experience with the development of a real-time observational registry fully integrated into the process of routine clinical workflow and show how this has the potential to transform research and quality assurance. Materials and Methods: From May 2011 to May 2016, physicians prospectively inputted data during the process of routine charting on patients seen in clinic. Using a customized interface established between an in-house registry and a commercially available, hospital-based electronic medical record system (Epic Systems, Verona, WI), a departmentally based parser was created for automatic data deposition, which was also linked to the Aria Treatment Planning Station (Varian Medical Systems, Palo Alto, CA). The total number of data fields embedded per disease site ranged from nine to 73 (median, 21 fields). Results: A total of 12,341 patients were logged into the registry, of whom 6,911 completed a course of radiation therapy. Primary disease sites were prostate (n = 2,340), breast (n = 2,159), head or neck (n = 1,426), primary CNS (n = 1,338), lung (n = 749), brain metastasis (n = 739), GI (n = 638), gynecologic (n = 534), and other or benign (n = 3,618). A total of 54 independent, investigator-initiated research studies have been initiated using queries supported by the registry from multiple access points, of which 23 were published in peer-reviewed journals. Conclusion: The development of a radiation oncology-specific registry enhanced research efficiency and facilitated quality assurance by producing clear and quality information to guide clinical practice.

Caglayan, Caglar; Terawaki, Hiromi; Chen, Qiushi; Rai, Ashish; Ayer, Turgay; Flowers, Christopher R.

2018 JCO Clinical Cancer Informatics

doi: 10.1200/CCI.17.00029pmid: 30652551

Purpose: Microsimulation is a modeling technique that uses a sample size of individual units (microunits), each with a unique set of attributes, and allows for the simulation of downstream events on the basis of predefined states and transition probabilities between those states over time. In this article, we describe the history of the role of microsimulation in medicine and its potential applications in oncology as useful tools for population risk stratification and treatment strategy design for precision medicine. Methods: We conducted a comprehensive and methodical search of the literature using electronic databases-Medline, Embase, and Cochrane-for works published between 1985 and 2016. A medical subject heading search strategy was constructed for Medline searches by using a combination of relevant search terms, such as "microsimulation model medicine," "multistate modeling cancer," and "oncology." Results: Microsimulation modeling is particularly useful for the study of optimal intervention strategies when randomized control trials may not be feasible, ethical, or practical. Microsimulation models can retain memory of prior behaviors and states. As such, it allows an explicit representation and understanding of how various processes propagate over time and affect the final outcomes for an individual or in a population. Conclusion: A well-calibrated microsimulation model can be used to predict the outcome of the event of interest for a new individual or subpopulations, assess the effectiveness and cost effectiveness of alternative interventions, and project the future disease burden of oncologic diseases. In the growing field of oncology research, a microsimulation model can serve as a valuable tool among the various facets of methodology available.