Interrater Reliability in Toxicity Identification: Limitations of Current Standards

Immune checkpoint inhibitors in the treatment of pleural mesothelioma: insights from real-world data

BACKGROUND

Immune checkpoint inhibitors (ICIs) have recently emerged as a promising strategy for the treatment of pleural mesothelioma (PM).

METHODS

This retrospective study evaluated treatment efficacy and safety in Japanese patients with PM treated with nivolumab and ipilimumab (N + I group: 41 patients) as first-line therapy and nivolumab monotherapy (N group: 33 patients) as second- or later-line treatment.

RESULTS

The median overall survival (OS) and progression-free survival (PFS) were not reached and 10.4 months in the N + I group, and 8.6 months and 3.5 months in the N group, respectively. Treatment-related adverse events (TRAEs) of any grade occurred in 68.3% of the N + I group and 72.7% of the N group, with grade 3-4 TRAEs in 19.5% and 12.1% of patients, respectively. Patients with an ECOG PS 0-1 had significantly better OS and PFS in both treatment groups (p < 0.001). In the N + I group, OS was significantly better in patients with TRAEs (p = 0.020) and in those with the epithelioid subtype (p = 0.047), although PFS was not significantly different (p = 0.138 and p = 0.154, respectively). In the N group, both OS (p = 0.007) and PFS (p = 0.048) were significantly longer in patients with TRAEs.

CONCLUSION

This study provides valuable real-world clinical evidence of the efficacy and safety of nivolumab plus ipilimumab and nivolumab monotherapy in Japanese patients with PM. These results support the use of ICIs as a viable treatment option for advanced or relapsed disease.

Managing strategies of chemotherapy and radiotherapy-induced oral mucositis

Radiotherapy and chemotherapy are widely employed as primary non-surgical cancer treatments; however, their non-selective cytotoxicity often leads to adverse events such as oral mucositis (OM), particularly in head and neck cancer therapies. International guidelines provide recommendations for managing chemoradiotherapy-induced OM in various clinical contexts. Subsequently, emerging researches have introduced evidence supporting novel approaches or existing regimens for OM prevention and treatment. The repurposing of established drugs has garnered significant interest due to its shorter development timeline, improved safety profiles, and lower costs compared to new drug development. For example, clinical trials assessing established drugs such as melatonin, clonidine, and pentoxifylline indicate promising potential for managing OM. Additionally, several emerging pharmacological interventions have demonstrated considerable efficacy; SAMITAL and rhIL-11 are supported by phase II clinical trials and prospective studies, while probiotics like Streptococcus salivarius K12 and curcumin have shown effectiveness in randomized clinical trials. Furthermore, recent high-level studies have reinforced the efficacy of non-pharmacological interventions, such as photobiomodulation (PBM) and cryotherapy, over the past two years. In all, given the evidence supporting different strategies, PBM and oral cryotherapy are highly recommended for managing OM when feasible. Topical clonidine, melatonin, oral pentoxifylline, topical SAMITAL or rhIL-11, oral SsK12, and curcumin may also be utilized but would benefit from validation in larger trials. Besides, Verbascoside, Palifermin, Amifostine, and Avasopasem manganese can be suggested for OM management, while the side effects should be monitored. The accessibility and cost/effectiveness of specific managing strategies of OM should be considered when selecting appropriate options.

Targeted RT study: results on early toxicity of targeted therapies and radiotherapy

PURPOSE/OBJECTIVE

Currently, there are few prospective data on the tolerability of combining targeted therapies (TT) with radiation therapy (RT). The objective of this prospective study was to assess the feasibility and toxicity of pairing RT with concurrent TT in cancer patients. The aim was to enhance the existing evidence base for the simultaneous administration of targeted substances together with radiotherapy.

METHODS

Prospective study enrollment was conducted at a single institution between March 1, 2020, and December 31, 2021, for all patients diagnosed with histologically confirmed cancer who underwent external beam radiotherapy in combination with targeted therapy. The study, known as the "targeted RT study," was registered in the German Clinical Trials Register under DRKS00026193. Systematic documentation of the toxicity profiles of different targeted therapies was performed, and the assessment of acute toxicity followed the guidelines of the National Cancer Institute Common Terminology Criteria for Adverse Events Version v5.0.

RESULTS

A total of 334 patients underwent 683 radiation therapy series. During the course of RT, 51 different TT substances were concurrently administered. External beam radiotherapy was employed for various anatomical sites. The combination of RT and concurrent TT administration was generally well tolerated, with no instances of severe acute toxicity observed. The most commonly reported toxicity was fatigue, ranging from mild to moderate Common Terminology Criteria for Adverse Events (CTCAE) °I-°III. Other frequently observed toxicities included dermatitis, dyspnea, dysphagia, and dry cough. No toxicity greater than moderate severity was recorded at any point. In only 32 patients (4.7% of evaluated RT series), the concurrent substance administration was discontinued due to side effects. However, these side effects did not exceed mild severity according to CTCAE, suggesting that discontinuation was a precautionary measure. Only one patient receiving Imatinib treatment experienced a severe CTCAE °III side effect, leading to discontinuation of the concurrent substance due to the sudden occurrence of melaena during RT.

CONCLUSION

In conclusion, the current study did not demonstrate a significant increase or additional toxicity when combining radiotherapy and concurrent targeted therapy. However, additional research is required to explore the specific toxicity profiles of the various substances that can be utilized in this context.

TRIAL REGISTRATION NUMBER

DRKS00026193. Date of registration 12/27/2022 (retrospectively registered).

Bleeding assessment following central venous catheter placement, a direct comparison of prospective and retrospective analyses

BACKGROUND

Reported bleeding incidences following central venous catheter (CVC) placement highly depend on methods of bleeding assessment. To determine the direction and magnitude of the bias associated with retrospective data collection, we used data from the PACER randomized controlled trial and a previous retrospective cohort study.

STUDY DESIGN AND METHODS

A patient-level comparison of CVC-related bleeding severity was made among (1) the prospectively collected clinical bleeding assessment of the PACER trial, (2) centralized assessment of CVC insertion site photographs, and (3) retrospective chart review. Interrater reliability for photographic bleeding assessment and retrospective chart review was assessed using Cohen's κ. The magnitude of underreporting of both methods compared to prospective clinical bleeding assessment at different cutoff points of clinically relevant bleeding was assessed using McNemar's test.

RESULTS

Interrater reliability was acceptable for both methods (κ = 0.583 and κ = 0.481 for photographic assessment and retrospective chart review, respectively). Photographic bleeding assessment led to significant underreporting of bleeding complications at all cutoff points. Retrospective chart review led to significant underreporting of minor bleeding complications, with an odds ratio (95% CI) of 0.17 (0.044-0.51) for the cutoff point grade 1 (i.e., self-limiting or requiring at most 20 min of manual compression) or higher. There was no significant underreporting of major bleeding complications with retrospective chart review.

DISCUSSION

Centralized photographic bleeding assessment and retrospective chart review lead to biased bleeding assessment compared to prospective clinical bleeding assessment.

Radiotherapy enhances efficacy of PD-1 inhibitors in advanced hepatocellular carcinoma: A propensity-matched real-world study

BACKGROUND

To address the need for immunotherapy in patients with advanced primary hepatocellular carcinoma (HCC), combination with radiotherapy (RT) has emerged as a promising strategy. In preclinical studies, irradiated tumors released tumor antigens to synergistically increase the antitumor effect of immunotherapy. Hence, we investigated whether RT enhances the efficacy of anti-programmed death receptor-1 (PD-1) inhibitors in advanced HCC in real-world practice.

METHODS

Between August 2018 and June 2021, 172 patients with advanced primary HCC were enrolled in the tertiary center (Zhongshan Hospital of Fudan University); 95 were treated with a combination of RT and the inhibitor of PD-1 (RT-PD1 cohort), and 77 were administered anti-PD-1 therapy (PD1 cohort). The first cycle of PD-1 inhibitors was administered within 60 days or concurrently with RT. Propensity score matching for bias reduction was used to evaluate the clinical outcomes.

RESULTS

Among 71 propensity-matched pairs, median progression-free survival was 5.7 months in the RT-PD1 cohort vs. 2.9 months in the PD1 cohort ( P  <0.001). Median overall survival was 20.9 months in the RT-PD1 cohort vs. 11.2 months in the PD1 cohort ( P  = 0.018). Compared with patients in the PD1 cohort, patients in the RT-PD1 cohort had significantly higher objective response rates (40.8%, 29/71 vs. 19.7%, 14/71, P  = 0.006) and disease control rates (62.0%, 44/71 vs. 31.0%, 22/71, P  <0.001). The incidences of toxic effects were not significantly different between the two cohorts.

CONCLUSIONS

RT plus anti-PD-1 therapy is well tolerated. RT enhances the efficacy of anti-PD-1 therapy in patients with advanced primary HCC by improving survival outcomes without increased toxic effects.

A potential treatment option for transformed small-cell lung cancer on PD-L1 inhibitor-based combination therapy improved survival

OBJECTIVES

Transformed small-cell lung cancer (T-SCLC) has an extremely poor prognosis, and no remedies based on immunotherapy have been evaluated among T-SCLC patients. We retrospectively analysed the efficacy and safety of combining atezolizumab with chemotherapy for T-SCLC.

METHODS

Forty-seven patients harbouring EGFR mutations who developed T-SCLC were enrolled. Eleven patients who used immunotherapy were defined as the I/O group, and the remaining 36 were defined as the Non-I/O group. Clinical characteristics, pathological data, and survival outcomes were collected. RNA sequencing and whole-exome sequencing (WES) were performed for in-depth analysis.

RESULTS

All patients received at least one line of EGFR-TKI before rebiopsy to confirm T-SCLC. Nine patients received atezolizumab-bevacizumab-carboplatin-paclitaxel (albumin-bound) (ABCP), and the remaining 2 received atezolizumab-etoposide-carboplatin (ECT) in the I/O group. The objective response rate was 73 % (8/11). The median progression-free survival (mPFS) of T-SCLC on post-transformation therapy with I/O group and Non-I/O group was 5.1 m and 4.1 m, respectively. The median post-T-SCLC overall survival of the I/O group was significantly longer than that Non-I/O group (20.2 m vs 7.9 m, P ＜ 0.01). T-SCLC harbouring EGFR L858R tended to be longer than EGFR 19del (mPFS: not reached vs 3.7 m, P = 0.11). Positive PD-L1 status was also associated with PFS benefits (mPFS: 6.0 m vs 3.7 m, P = 0.20). Furthermore, RNA sequencing revealed that expression of SFTPA1 is significantly higher in the durable clinical benefit group. WES showed that STC2 mutation is more frequently observed at the time-point immunotherapy acquired resistance. Combination therapy based on a PD-L1 inhibitor was well tolerated, and the safety profile was consistent with previously reported studies.

CONCLUSION

Our study first demonstrated that a PD-L1 inhibitor combined with chemotherapy ± bevacizumab could be a potential safe option for specific SCLC-transformed patients. Subsequent studies with more patients are essential to verify the efficacy and potential biomarkers.

Safety and Validity of Anterior Cervical Disc Replacement for Single-level Cervical Disc Disease: Initial Two-year Follow-up of the Prospective Observational Post-marketing Surveillance Study for Japanese Patients

Anterior cervical disc replacement (ACDR) using cervical artificial disc (CAD) has the advantage of maintaining the range of motion (ROM) at the surgical level, subsequently reducing the postoperative risk of adjacent disc disease. Following the approval for the clinical use in Japan, a post-marketing surveillance (PMS) study was conducted for two different types of CAD, namely, Mobi-C (metal-on-plastic design) and Prestige LP (metal-on-metal design). The objective of this prospective observational multicenter study was to analyze the first 2-year surgical results of the PMS study of 1-level ACDR in Japan. A total of 54 patients were registered (Mobi-C, n = 24, MC group; Prestige LP, n = 30, PLP group). Preoperative neurological assessment revealed radiculopathy in 31 patients (57.4%) and myelopathy in 15 patients (27.8%). Preoperative radiological assessment classified the disease category as disc herniation in 15 patients (27.8%), osteophyte in 6 patients (11.1%), and both in 33 patients (61.1%). The postoperative follow-up rates at 6 weeks, 6 months, 1 year, and 2 years after ACDR were 92.6%, 87.0%, 83.3%, and 79.6%, respectively. In both groups, patients' neurological condition improved significantly after surgery. Radiographic assessment revealed loss of mobility at the surgical level in 9.5% of patients in the MC group and in 9.1% of patients in the PLP group. No secondary surgeries at the initial surgical level and no serious adverse events were observed in either group. The present results suggest that 1-level ACDR is safe, although medium- to long-term follow-up is mandatory to further verify the validity of ACDR for Japanese patients.

Deep Learning for Cancer Symptoms Monitoring on the Basis of Electronic Health Record Unstructured Clinical Notes

PURPOSE

Symptoms are vital outcomes for cancer clinical trials, observational research, and population-level surveillance. Patient-reported outcomes (PROs) are valuable for monitoring symptoms, yet there are many challenges to collecting PROs at scale. We sought to develop, test, and externally validate a deep learning model to extract symptoms from unstructured clinical notes in the electronic health record.

METHODS

We randomly selected 1,225 outpatient progress notes from among patients treated at the Dana-Farber Cancer Institute between January 2016 and December 2019 and used 1,125 notes as our training/validation data set and 100 notes as our test data set. We evaluated the performance of 10 deep learning models for detecting 80 symptoms included in the National Cancer Institute's Patient-Reported Outcomes version of the Common Terminology Criteria for Adverse Events (PRO-CTCAE) framework. Model performance as compared with manual chart abstraction was assessed using standard metrics, and the highest performer was externally validated on a sample of 100 physician notes from a different clinical context.

RESULTS

In our training and test data sets, 75 of the 80 candidate symptoms were identified. The ELECTRA-small model had the highest performance for symptom identification at the token level (ie, at the individual symptom level), with an F1 of 0.87 and a processing time of 3.95 seconds per note. For the 10 most common symptoms in the test data set, the F1 score ranged from 0.98 for anxious to 0.86 for fatigue. For external validation of the same symptoms, the note-level performance ranged from F1 = 0.97 for diarrhea and dizziness to F1 = 0.73 for swelling.

CONCLUSION

Training a deep learning model to identify a wide range of electronic health record-documented symptoms relevant to cancer care is feasible. This approach could be used at the health system scale to complement to electronic PROs.

Feasibility, pitfalls and results of a structured concept-development phase for a randomized controlled phase III trial on radiotherapy in primary prostate cancer patients

OBJECTIVE

Failure rate in randomized controlled trials (RCTs) is > 50%, includes safety-problems, underpowered statistics, lack of efficacy, lack of funding or insufficient patient recruitment and is even more pronounced in oncology trials. We present results of a structured concept-development phase (CDP) for a phase III RCT on personalized radiotherapy (RT) in primary prostate cancer (PCa) patients implementing prostate specific membrane antigen targeting positron emission tomography (PSMA-PET).

MATERIALS AND METHODS

The 1 yr process of the CDP contained five main working packages: (i) literature search and scoping review, (ii) involvement of individual patients, patients' representatives and patients' self-help groups addressing the patients' willingness to participate in the preparation process and the conduct of RCTs as well as the patient informed consent (PIC), (iii) involvement of national and international experts and expert panels (iv) a phase II pilot study investigating the safety of implementation of PSMA-PET for focal dose escalation RT and (v) in-silico RT planning studies assessing feasibility of envisaged dose regimens and effects of urethral sparing in focal dose escalation.

RESULTS

(i) Systematic literature searches confirmed the high clinical relevance for more evidence on advanced RT approaches, in particular stereotactic body RT, in high-risk PCa patients. (ii) Involvement of patients, patient representatives and randomly selected males relevantly changed the PIC and initiated a patient empowerment project for training of bladder preparation. (iii) Discussion with national and international experts led to adaptions of inclusion and exclusion criteria. (iv) Fifty patients were treated in the pilot trial and in- and exclusion criteria as well as enrollment calculations were adapted accordingly. Parallel conduction of the pilot trial revealed pitfalls on practicability and broadened the horizon for translational projects. (v) In-silico planning studies confirmed feasibility of envisaged dose prescription. Despite large prostate- and boost-volumes of up to 66% of the prostate, adherence to stringent anorectal dose constraints was feasible. Urethral sparing increased the therapeutic ratio.

CONCLUSION

The dynamic framework of interdisciplinary working programs in CDPs enhances robustness of RCT protocols and may be associated with decreased failure rates. Structured recommendations are warranted to further define the process of such CDPs in radiation oncology trials.

Whole brain radiotherapy combined with intrathecal liposomal cytarabine for leptomeningeal metastasis-a safety analysis and validation of the EANO-ESMO classification

BACKGROUND

Although there is no proven standard therapy for leptomeningeal metastases (LM), treatment often includes intrathecal chemotherapy combined with whole brain radiation therapy (WBRT). Little is known about the toxicity of such combination therapies. We performed a retrospective safety analysis for the combination of intrathecal liposomal cytarabine with WBRT in patients with LM and validated the EANO-ESMO (European Association of Neuro-oncology-European Society for Medical Oncology) classification in this unique cohort.

METHODS

Treatment toxicities in patients diagnosed with LM between 2004 and 2014 were retrospectively analyzed according to RTOG (Radiation Therapy Oncology Group) toxicity criteria and NCI CTCAE V5.0 (National Cancer Institute Common Terminology Criteria for Adverse Events, Version 5.0). Diagnostic criteria and treatment response as assessed by EANO-ESMO classification were correlated with survival by Kaplan-Meier analysis and Breslow test.

RESULTS

In all, 40 patients with LM who were treated with combined WBRT and intrathecal cytarabine, were identified. Ten patients (25%) experienced adverse events ≥grade 3 according to RTOG toxicity criteria; in 22 patients (55%) NCI CTCAE ≥grade 3 were detected. Median overall survival was 124 days. Median time to neurological progression was 52 days. Patients with positive cerebrospinal fluid (CSF) cytology (n = 26) showed worse prognosis compared to patients with negative CSF cytology (n = 14; mOS (median overall survival) 84 days versus 198 days, p = 0.006, respectively). The EANO-ESMO response assessment was significantly associated with survival: "stable" (n = 7) mOS 233 days, "response" (n = 10) mOS 206 days, "progression" (n = 17) mOS 45 days, "suspicion of progression" (n = 6) mOS 133 days; overall, p < 0.001.

CONCLUSIONS

In this retrospective analysis, combined treatment of WBRT and intrathecal liposomal cytarabine shows an acceptable safety profile and may indicate a trend towards improved efficacy. The EANO-ESMO classification for diagnosis and treatment response predicts survival.

Natural Language Processing to Identify Advance Care Planning Documentation in a Multisite Pragmatic Clinical Trial

CONTEXT

Large multisite clinical trials studying decision-making when facing serious illness require an efficient method for abstraction of advance care planning (ACP) documentation from clinical text documents. However, the current gold standard method of manual chart review is time-consuming and unreliable.

OBJECTIVES

To evaluate the ability to use natural language processing (NLP) to identify ACP documention in clinical notes from patients participating in a multisite trial.

METHODS

Patients with advanced cancer followed in three disease-focused oncology clinics at Duke Health, Mayo Clinic, and Northwell Health were identified using administrative data. All outpatient and inpatient notes from patients meeting inclusion criteria were extracted from electronic health records (EHRs) between March 2018 and March 2019. NLP text identification software with semi-automated chart review was applied to identify documentation of four ACP domains: (1) conversations about goals of care, (2) limitation of life-sustaining treatment, (3) involvement of palliative care, and (4) discussion of hospice. The performance of NLP was compared to gold standard manual chart review.

RESULTS

435 unique patients with 79,797 notes were included in the study. In our validation data set, NLP achieved F1 scores ranging from 0.84 to 0.97 across domains compared to gold standard manual chart review. NLP identified ACP documentation in a fraction of the time required by manual chart review of EHRs (1-5 minutes per patient for NLP, vs. 30-120 minutes for manual abstraction).

CONCLUSION

NLP is more efficient and as accurate as manual chart review for identifying ACP documentation in studies with large patient cohorts.

Minimally important differences for the EORTC QLQ-C30 in prostate cancer clinical trials

Background

The aim of the study was to estimate the minimally important difference (MID) for interpreting group-level change over time, both within a group and between groups, for the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) scores in patients with prostate cancer.

Methods

We used data from two published EORTC trials. Clinical anchors were selected by strength of correlations with QLQ-C30 scales. In addition, clinicians’ input was obtained with regard to plausibility of the selected anchors. The mean change method was applied for interpreting change over time within a group of patients and linear regression models were fitted to estimate MIDs for between-group differences in change over time. Distribution-based estimates were also evaluated.

Results

Two clinical anchors were eligible for MID estimation; performance status and the CTCAE diarrhoea domain. MIDs were developed for 7 scales (physical functioning, role functioning, social functioning, pain, fatigue, global quality of life, diarrhoea) and varied by scale and direction (improvement vs deterioration). Within-group MIDs ranged from 4 to 14 points for improvement and − 13 to − 5 points for deterioration and MIDs for between-group differences in change scores ranged from 3 to 13 for improvement and − 10 to − 5 for deterioration.

Conclusions

Our findings aid the meaningful interpretation of changes on a set of EORTC QLQ-C30 scale scores over time, both within and between groups, and for performing more accurate sample size calculations for clinical trials in prostate cancer.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-021-08609-7.

Clinical Natural Language Processing for Radiation Oncology: A Review and Practical Primer

Natural language processing (NLP), which aims to convert human language into expressions that can be analyzed by computers, is one of the most rapidly developing and widely used technologies in the field of artificial intelligence. Natural language processing algorithms convert unstructured free text data into structured data that can be extracted and analyzed at scale. In medicine, this unlocking of the rich, expressive data within clinical free text in electronic medical records will help untap the full potential of big data for research and clinical purposes. Recent major NLP algorithmic advances have significantly improved the performance of these algorithms, leading to a surge in academic and industry interest in developing tools to automate information extraction and phenotyping from clinical texts. Thus, these technologies are poised to transform medical research and alter clinical practices in the future. Radiation oncology stands to benefit from NLP algorithms if they are appropriately developed and deployed, as they may enable advances such as automated inclusion of radiation therapy details into cancer registries, discovery of novel insights about cancer care, and improved patient data curation and presentation at the point of care. However, challenges remain before the full value of NLP is realized, such as the plethora of jargon specific to radiation oncology, nonstandard nomenclature, a lack of publicly available labeled data for model development, and interoperability limitations between radiation oncology data silos. Successful development and implementation of high quality and high value NLP models for radiation oncology will require close collaboration between computer scientists and the radiation oncology community. Here, we present a primer on artificial intelligence algorithms in general and NLP algorithms in particular; provide guidance on how to assess the performance of such algorithms; review prior research on NLP algorithms for oncology; and describe future avenues for NLP in radiation oncology research and clinics.

Efficacy and safety analysis of dexamethasone-lipiodol emulsion in prevention of post-embolization syndrome after TACE: a retrospective analysis

Background

To investigate the efficacy and safety of dexamethasone-lipiodol emulsion in the prevention of post-embolization syndrome after TACE.

Method

The data of 255 patients who underwent TACE in the interventional department from June 2017 to June 2020 were collected. This is a retrospective assessment of patients who were non-randomly treated with dexamethasone in TACE. The patients were divided into two groups: TACE using lipiodol + chemotherapeutic emulsion group (TACE group, N = 133); TACE using lipiodol + dexamethasone + chemotherapeutic emulsion group (TACE + dexamethasone group, N = 122). Primary study endpoint: incidence of abdominal pain, fever, nausea and vomiting 0–72 h after TACE in both groups. Secondary study endpoints: incidence of infection after TACE in both groups.

Results

Incidence of post-embolization syndrome after TACE (TACE group vs TACE + dexamethasone group): abdominal pain, 55.6% versus 36.1% (P value 0.002); fever, 37.6% versus 13.1% (P value 0.000); nausea, 60.9% versus 41.0% (P value 0.001); vomiting, 48.1% versus 21.3% (P value 0.000). Incidence of infection after TACE (TACE group vs TACE + dexamethasone group): 1.5% versus 2.5% (P value 0.583).

Conclusion

The lipiodol + dexamethasone emulsion can significantly reduce the incidence rate of post-embolization syndrome after TACE, with exact effect and high safety.

Analysis of Patient-Reported Outcome Utilization Within National Clinical Trials Network Cooperative Group Radiation Oncology Trials Over the Past 2 Decades

PURPOSE

When treating cancer, both quantity and quality of life are valuable, though oncology trials have long placed greater emphasis on the former. The goal of this work was to evaluate how patient-reported outcomes (PROs) have been incorporated into radiation therapy trials within the National Clinical Trials Network over the last 2 decades to measure quality of life and to assess how PRO data have been disseminated in publications upon trial conclusion.

METHODS AND MATERIALS

This cross-sectional study analyzed the frequency of use of PROs in National Clinical Trials Network cooperative group radiation therapy phase 2 and 3 clinical trials over the past 2 decades. A literature review was performed to determine the publication outcomes of PRO data, including only trials that used PROs in their design and were mature enough to have published results.

RESULTS

Fifty-seven (56.4%) of the 101 trials included in this study included PROs in their design. Brain and head and neck trials demonstrated the largest proportional incorporation of PROs (81.8% and 76.9%, respectively), and thoracic and breast trials used the fewest (18.8% and 37.5%, respectively). The EQ-5D family of questionnaires was the most commonly used PROs, used in 22.8% of trials included. The literature review demonstrated a pattern of increased publication of PRO data alongside survival endpoints in manuscripts derived from these trials over time.

CONCLUSIONS

Though there is room for improvement, the field of radiation oncology has embraced the incorporation of PROs into multicenter, high-impact clinical trials over the past 2 decades and has increased its publication of this data alongside survival data from these trials.

Natural language processing for abstraction of cancer treatment toxicities: accuracy versus human experts

Objectives

Expert abstraction of acute toxicities is critical in oncology research but is labor-intensive and variable. We assessed the accuracy of a natural language processing (NLP) pipeline to extract symptoms from clinical notes compared to physicians.

Materials and Methods

Two independent reviewers identified present and negated National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) v5.0 symptoms from 100 randomly selected notes for on-treatment visits during radiation therapy with adjudication by a third reviewer. A NLP pipeline based on Apache clinical Text Analysis Knowledge Extraction System was developed and used to extract CTCAE terms. Accuracy was assessed by precision, recall, and F1.

Results

The NLP pipeline demonstrated high accuracy for common physician-abstracted symptoms, such as radiation dermatitis (F1 0.88), fatigue (0.85), and nausea (0.88). NLP had poor sensitivity for negated symptoms.

Conclusion

NLP accurately detects a subset of documented present CTCAE symptoms, though is limited for negated symptoms. It may facilitate strategies to more consistently identify toxicities during cancer therapy.