Trends in patient-reported outcome use in early phase dose-finding oncology trials - an analysis of ClinicalTrials.gov.

BACKGROUND

The primary purpose of early phase (phase 1 and phase 1/2) oncology trials is to establish the safety and tolerability of novel anti‐cancer agents. Clinician‐assessed adverse event (AE) reporting using the Common Terminology Criteria for Adverse Events (CTCAE) is critical for describing a drug's safety profile. CTCAE data, in conjunction with data on dose modifications, discontinuations and hospitalisations, is then used to select tolerable doses and regimens, including the maximum tolerated dose (MTD) and recommended phase 2 dose (RP2D).

However, clinician‐assessed AE reporting has limitations. Firstly, it may miss up to half of symptomatic AEs compared with patient self‐reports. 1 , 2 , 3 , 4 , 5 Some AEs may be difficult to observe (e.g., fatigue) and not adequately characterised. This may under‐estimate toxicities and result in suboptimal toxicity management. Secondly, the patient's perspective on AEs is noticeably absent in the selection of tolerable doses and regimens.

Changes in the types of therapies in early phase oncology trials can also produce challenges in defining tolerable doses. Immunotherapy and targeted therapy may not produce toxicities in a dose‐dependent manner. Toxicities may be longer in duration or occur later than the traditional dose‐limiting toxicity (DLT) period, which is often defined as the first cycle of treatment. 6 , 7 Low‐grade toxicities experienced over the medium term to long term could reduce a drug's tolerability but are not taken into consideration. This can make defining the MTD, and consequently RP2D, challenging. Alternative methods of determining tolerable doses and regimens are, therefore, urgently needed.

There is increasing interest from clinicians, industry, and regulators, including the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA), to enhance the accuracy of toxicity reporting and incorporate the patient's experience in the drug development process through the use of patient‐reported outcomes (PROs). 8 , 9 , 10

PROs are defined as measurements of a patient's health status that come directly from the patient. 11 PROs have been extensively studied in the clinical trial and routine care settings and are reliable, feasible and valued by clinicians and patients. 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 PROs, in conjunction to CTCAE data, could provide different yet complementary data to inform a more complete understanding of a drug's tolerability and support the selection of tolerable doses and schedules. 21 PRO collection in early phase trials may become more crucial as drugs developed for specific molecular subtypes with small patient populations may receive regulatory approval on the basis of early phase trials resulting in PROs not being collected in later‐phase trials. A recent example is the FDA approval of the neutrotrophic tyrosine receptor kinase (NTRK) inhibitor entrectinib in patients with NTRK gene fusion solid tumours or ROS1 positive metastatic non‐small‐cell‐lung cancer. 22

The FDA has taken steps towards the inclusion of the patient perspective in all stages of the drug development process, including issuing guidance on the use of PROs in drug development, and collaborating with industry to form the PRO consortium with the aim of developing robust patient‐reported symptom measurement tools. 11

Despite the desire to increase PRO use in drug development, there is minimal literature regarding the use of PROs in early phase oncology trials. A review of PRO use across all clinical trials (including non‐oncological trials) from 2007 to 2013 showed phase 1 trials were less likely to include a PRO measure (PROM) compared with phase 2 trials. 23 A systematic review from 2012 to 2016 identified 15 phase 1 oncology trials with health‐related quality of life (HRQOL) as an endpoint, none of which used HRQOL to inform the RP2D. 24 However, this study only included published trials, placing it at risk of publication bias. PRO inclusion is also a relatively recent phenomena, so examining published trials alone may reflect choices made by trial investigators from several years prior. Trials in progress may have higher rates of PRO inclusion due to increasing awareness of PROs among early phase trialists.

The feasibility and importance of assessing PROs was highlighted in a single‐centre study in which 80 PRO‐CTCAE items were assessed in phase 1 oncology patients at baseline, mid‐cycle 1 and 2. 25 , 26 , 27 Completion rates were high (98.7%). Significant under‐reporting of certain AEs by clinicians was noted, including fatigue, anxiety, and pain. There was also poor patient‐clinician agreement on sexual health, cognition, and urination. This emphasised the potential role of PROs in complementing clinician‐assessed CTCAE gradings to formulate a more comprehensive understanding of a drug's tolerability. PRO‐CTCAE responses, particularly interference, have clinically relevant consequences and correspond with DLTs, dose interruption, and drug discontinuation. This provides further evidence of their usefulness in assessing drug tolerability. 28

We conducted a database search of ClinicalTrials.gov to describe characteristics and trends of PRO use and reporting in early phase oncology trials. ClinicalTrials.gov is the largest global trial registry and a valuable resource for assessing the trends and characteristics of registered clinical trials. 29 By including trials that are completed and in progress, this will provide a more current and complete assessment of PRO use.

METHOD

Study design and search strategy

This cross‐sectional study included early phase dose‐finding oncology trials (DFOT) with start dates 01/01/2007–20/01/2020 registered on ClinicalTrials.gov. Ethics approval was not required for this database review of ClinicalTrials.gov. Data were extracted in XML format on 1 February 2020. DFOT were identified using the following search strategy in the advanced search function: condition or disease – cancer; study type – interventional (clinical trial); study results – all trials; recruitment status – not yet recruiting, recruiting, enrolling by invitation, active not recruiting, suspended and completed; expanded access – available, no longer available, temporarily not available, approved for marketing; age: child, adult, older adult; sex – all; study phase – early phase 1, phase 1; funder type – NIH, other US federal agency, industry, all others; date restriction (01/01/2007–20/01/2020). The search was then repeated to identify DFOT with a PRO endpoint by adding outcome measures – ‘patient‐reported outcome’ OR ‘Health‐related quality of life’ OR ‘quality of life’ OR ‘QOL’ OR ‘PRO’ OR ‘PROMS’ OR ‘HRQOL’.

The ClinicalTrials.gov database entry for each search result was reviewed to confirm inclusion criteria were met. Eligible trials were DFOT (phase 1 or phase 1/2 oncology trial with a dose finding component) assessing an intervention (drug or radiotherapy), which included at least one PRO as an endpoint. Haematology and paediatric trials were included. Trials that did not meet all eligibility criteria were excluded.

Data extraction

Data was extracted using a pre‐defined data abstraction form in Microsoft Excel (Text S1 ). Study characteristics were extracted: title, study period, sponsor's country of origin, sponsor type, number of participating centres, number of patients enrolled, study phase, dose escalation study design, tumour type, type of therapy undergoing dose escalation, primary endpoint for phase 1, and current study activity status. PRO characteristics were then extracted: number of PROs included, type of PROs, PRO endpoint (primary, secondary, exploratory), phase in trial in which PROs were collected, method of collection, person completing the PRO, frequency of PRO assessment and duration of PRO follow‐up.

Patient‐reported outcome reporting

Data were extracted regarding PRO outcomes and the statistical methods used for PRO analysis from two sources: Completed trials with PRO results available on the ClinicalTrials.gov database Trials with published manuscripts listed on the ClinicalTrials.gov database.

There are no guidelines regarding how PRO data should be reported in non‐randomised trials. The following details were extracted based on the CONSORT‐PRO checklist for randomised trials 30 : whether PRO results were published, whether PRO results were described in the abstract, whether a PRO hypothesis was stated, whether relevant PRO domains were identified, the content validity of the PRO, timing of PRO assessments, statistical methods for PRO analysis, presence of missing data and reasons for missing data, handling of missing data, whether a minimal clinically important difference (MCID) was described and whether PRO data informed the MTD or RP2D.

All data were checked for internal consistency and disagreements resolved by discussion among the investigators.

Data analysis

Data were analysed using R version 3.6.1. For continuous variables, summary statistics of median and range were displayed. For categorical variables, frequencies (number) and percentages were displayed. Linear regression models with ‘year of study initiation’ as the independent variable were fitted to assess the trend of the number and percentage of trials, and number of sponsoring countries with PRO endpoints over time.

RESULTS

A total of 10 372 DFOTs were identified on ClinicalTrials.gov. In total, 548 (5.3%) DFOTs included a PRO endpoint, and 231 (42.2%) met eligibility criteria (Figure 1 ): adult (224, 97%), paediatric (7, 3%), solid tumour (175, 75.8%) and haematology (56, 24.2%) (Table 1 ). Phase 1/2 (108, 46.8%) trials were most common, followed by phase 1 dose escalation (101, 43.7%) and phase 1 dose escalation and expansion trials (22, 9.5%).

All eligible trials

Study characteristics

The number and percentage of DFOT with PRO endpoints increased by 2.3 trials per year (95% confidence interval (CI): 1.6–2.9) (Figure 2A ) and 0.3% per year (95% CI: 0.2–0.4) (Figure 2B ) respectively. There was also an increasing number of countries with institutions sponsoring DFOT with PRO endpoints, from three countries in 2007 to 11 in 2019 (Figure 3A ), at a rate of 0.7 countries per year (95% CI: 0.4–0.9) (Figure 3B ).

Trials were predominantly conducted in adults (224, 97.0%) with solid tumours (175, 75.8%), sponsored by individuals, universities or other organisations (135, 58.4%). A similar number involved drug combinations (119, 51.5%) or single agents (110, 47.6%). The most common therapies in escalation were targeted therapy (94, 40.7%), immunotherapy (33, 14.3%) and radiotherapy (33, 14.3%). Most trials did not specify their study design (151, 65.4%). 64 (27.7%) used a ‘3+3’ dose escalation design and 12 (5.2%) used a continual reassessment method. MTD (107, 34.9%) and safety (95, 30.9%) were the most common primary endpoints. A total of 114 (49.4%) were still recruiting, and 77 (33.3%) were completed.

Patient‐reported outcome characteristics

PROs were typically secondary endpoints (207, 89.6%). Most trials (137, 59.3%) used 1 PROM (range: 1–7). PROs were most frequently collected in the dose escalation (115, 49.8%) and phase 1 and 2 (54, 23.4%). A total of 119 unique PROMs were used. The most common were the EORTC QLQ C30 (78, 21.1%), EQ‐5D‐5L (19, 5.1%), the Brief Pain Inventory (10, 2.7%) and PRO‐CTCAE (10, 2.7%). Information on the frequency of PRO assessment was missing in 140 (60.6%) trials, and the method of PRO collection was missing in 230 (99.6%) trials. Only 13 trials (5.6%) specified the person completing the PRO.

Eligible trials with reported results on ClinicalTrials.gov

Fifteen completed trials had results available on ClinicalTrials.gov database, of which eight reported their PRO results specifically (Figure 1 ). The study characteristics and PRO reporting characteristics are presented in Table S1 . Seven trials presented descriptive statistics, and one presented least‐square means of treatment effects from a linear mixed effects (LME) model.

Eligible trials with published manuscripts listed on ClinicalTrials.gov

Study characteristics

Eighteen trials had published manuscripts listed on the ClinicalTrials.gov database (Table 2 ). Trials were primarily sponsored by individuals, universities or other organisations (9, 50.0%). Most used a ‘3+3’ dose escalation design (10, 55.6%). The treatments most common in dose escalation were targeted therapy (7, 38.9%) and chemotherapy (6, 33.3%). All trials except one had safety and/or MTD as their primary endpoint. 31

Three trials also listed their PRO results on the ClinicalTrials.gov database. As expected, more detailed descriptive and inferential analyses were provided in the manuscripts compared with the entries on the ClinicalTrials.gov database.

PRO reporting

One trial ( NCT00692900 ) conducted the dose escalation in two parallel arms and reported each arm separately in two publications. 32 , 33 This resulted in 19 publications from 18 trials (Table 3 ). 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 Only one trial included PROs as a co‐primary endpoint. 34 No trials stated a PRO hypothesis. All trials used a valid PRO instrument. The most commonly used PROMs were the EORTC QLQ C30 (7, 26.9%), FACT‐hepatobiliary (2, 7.7%), International Prostate Symptom Score (2, 7.7%) and the Myelofibrosis Symptom Assessment Form (2, 7.7%). The majority collected PROs during the Phase I dose escalation (11, 61.1%). Most (16, 88.9%) collected PROs at baseline and subsequent time points. No trials identified who was responsible for completing the PROs or how PROs were collected.

The statistical methods for PRO analysis were variably described. Analysis approaches were classified into three categories: descriptive statistics, inferential statistics (using the observed data to draw inferences, generalise and make judgements about the larger population, usually via hypothesis testing with p ‐values) or a combination. Nine (50.0%) used descriptive statistics, 1 (5.6%) used inferential statistics, 5 (27.8%) used both. Three (16.7%) did not provide sufficient details on their statistical methods to enable classification. Trials using descriptive analysis described changes in PRO scores over time using different categories, such as those whose PRO scores improved, remained stable or declined. 35 , 41 , 44 , 45 Among those trials using inferential statistics, most conducted simple statistical tests such as the t ‐test to assess whether there were significant differences in PRO scores between baseline and after start of treatment, or among different treatment arms. One trial used a Cox regression model to compare time of worsening of symptom scores between two randomised treatment groups in the phase 2 component and presented the estimated hazard ratio and CI. Another trial used an LME model to account for repeated measures over time and allowed for missing data at random. Least‐square means of the treatment effects were obtained from an LME model (with adjustment for baseline PRO scores) to assess whether the experimental treatment had a significant effect on PROs over time compared with placebo in its exploratory randomised controlled component. 34 Sixteen (88.9%) analysed PRO data over time, four (22.2%) analysed PRO data over different doses. Four (22.2%) trials described an MCID. Nine (50.5%) reported the number of patients with missing data, and two (11.1%) reported the reasons for missing data. Two (11.1%) described their methods for dealing with missing data.

With regards to using PROs to define tolerable doses and regimens, 3 (16.7%) used PROs to confirm the MTD. However, these do not state whether PROs were reviewed prior to determining the MTD or whether the MTD was retrospectively deemed tolerable after PRO data were reviewed.

Seven (39.8%) trials included PRO results in their abstract. All trials reported their PRO results in their primary trial manuscript rather than in a separate manuscript.

DISCUSSION

This is the first study to examine current trends in PRO use and reporting in DFOT. PRO use in DFOT increased over time and in a wider variety of settings. The trial characteristics are representative of the current drug development landscape, with a trend towards combinations of therapies in seamless phase 1/2 trials with novel dose‐finding statistical designs such as the Continual Reassessment Method. 50 , 51 , 52 , 53 Around half (58.4%) were sponsored by individuals, universities or other organisations, primarily representing academic institutions. This is nearly double the rate of academic sponsorship of DFOT on ClinicalTrials.gov for the same time period (2628/10372, 25.3%), indicating the key role that academic‐sponsored trials play in driving the inclusion of patient‐centred endpoints.

Nevertheless, overall use remains limited. DFOT including PROs remain a small proportion of all DFOT (Figure 2B ), and this only increased by 2.3 trials per year (Figure 2A ). PROs were predominantly used as a secondary or exploratory endpoint, indicating that researchers are not using PROs for the primary endpoint of dose determination. Generic cancer or disease‐specific PROMs were mainly used. However, these may not adequately capture the breadth of toxicities a patient may experience on DFOT and may not be fit for the purpose of assessing tolerability. Some trials used item libraries such as the PRO‐CTCAE (10, 2.7%) and EORTC item library (1, 0.3%), which may be useful for examining specific toxicities from novel agents not covered by legacy measures. Further work is needed to determine how best to select these items. A balance will need to be struck between thoroughness (such as including all 78 symptomatic toxicities in the PRO‐CTCAE to ensure unexpected toxicities are not missed) and feasibility (whether patients will complete lengthy PROMs on a weekly basis). 54 A possible compromise 55 is to collect a core set of items from an item library representing common and clinically relevant treatment‐related symptoms 56 and anticipated toxicities from pre‐clinical trials, 57 with a free text item to ensure unexpected toxicities are not missed. 58

The number of trials reporting PRO results was also limited. Only 8 completed trials had PRO results available on the ClinicalTrials.gov database, and only 18 trials published their PRO results in a manuscript listed on ClinicalTrials.gov, indicating a significant amount of research waste. This is also an issue in later‐phase trials with around 20% reporting their PRO results. 59 , 60

Although there are no guidelines to inform how PRO results should be reported for non‐randomised trials, the CONSORT‐PRO extension provides a useful checklist of items that should be reported to enable interpretation of PRO results. Key information such as a PRO objective, the person completing the PRO, and the method of PRO collection was not described in any publications. Furthermore, the statistical approach to the PRO data, MCID, the presence of missing data (and the reasons for missing data) were only described in some publications. This limits the interpretation and usefulness of the PRO data for making decisions regarding tolerability. It also limits the capacity to develop hypotheses and inform power calculations for later‐phase trials. This is again consistent with the later‐phase trial setting, where studies have shown that PRO reporting quality is suboptimal. 61 , 62 , 63 , 64

Although most trials collected PRO data over different doses and over several time points, only 22.2% analysed PRO data over different doses and time. More than 60% reported PRO data over time only. This fails to use the richness of the data to further inform tolerability at different doses. In addition to descriptive summary measures, several regression approaches can be used to analyse PRO data over time and different doses such as LME model or generalised estimating equation framework, which can account for the correlation of PRO measurements at different time points within an individual. 65 , 66 Only a small proportion (16.7%) used PROs to confirm the MTD. It is unclear how the PRO information was incorporated to inform the MTD, and whether it was reviewed prior or after the determination of MTD.

The strengths of this study are that by searching ClinicalTrials.gov rather than published trials we generated a current and complete assessment of PRO use. Many DFOT are never published or there can be a significant lag time between completion and publication. 67 , 68 Therefore, searching for published trials only may provide a limited picture of PRO use. We included all forms of DFOT, including haematology, paediatric and radiotherapy trials, which have been excluded from prior reviews. 24 Our search strategy included all trials commencing from 2007 when registration of US‐based trials to ClinicalTrials.gov became compulsory, ensuring this review is as current as possible. However, not all trials sponsored outside the United States are registered on ClinicalTrials.gov, which may result in those trials being under‐represented in our study.

Other limitations include the use of the ClinicalTrials.gov advanced search function, which can return slight variation in the number of search results day‐to‐day. This was mitigated by creating a copy of the database prior to data extraction. Our search strategy did not include specific PROMs, which may have resulted in some DFOT being missed. However, given that 119 unique PROMs were identified using our search strategy, it is likely that the vast majority of DFOT using PROs were captured by using generic QOL‐ and PRO‐related terms. There was also significant variability in the quality of data provided for each trial. We cannot be certain these data are accurate as we did not review the individual trial protocols or published manuscripts. However, the Clinical Trials Transformation Initiative suggests that information is more likely to be complete if the trial is evaluating drugs/devices that are sponsored by US or multinational organisations which the majority of these trials were. 69 Although inaccuracies and missing data may impact on the quality and characteristics of the trials, the purpose of this study was to examine trends and characteristics in PRO use. This is more likely to introduce random error rather than bias as the issues are likely to be distributed similarly across time.

The evaluation of PRO reporting quality was not performed as a systematic review. There may be trials which have published their results but have not updated their records on ClinicalTrials.gov. However, there were several similarities noted with our study and the systematic review by Fiteni et al., 24 including the type of PROMs used and inadequacies in PRO result reporting, reinforcing that our findings are a fair representation of the characteristics and quality of PRO reporting in DFOT.

Further work is necessary to determine the degree of acceptability of PROs to key stakeholders, potential benefits and barriers and the role of PRO data in defining tolerable doses and regimens. A qualitative study examined patient and clinician attitudes towards the collection of electronic PROs in early phase oncology trials. 70 While most agreed PROs could provide a more comprehensive understanding of a drug's toxicity, clinicians expressed concerns about monitoring PRO data and the need for careful decision‐making regarding data flow and symptom attribution. Our group is currently conducting a global survey exploring these questions in clinicians, statisticians, trial managers, funders and regulators.

Adaptation of existing methodological guidelines for the inclusion of PROs in randomised trials (e.g. SPIRIT‐PRO extension, CONSORT‐PRO extension, SISAQOL) 30 , 71 , 72 will be needed to standardise protocol content, reporting and statistical analysis in early phase trials. Guidance from regulators such as the EMA and FDA will also enhance uptake. 11 Simultaneous publication of PRO results alongside the main trial manuscript should be encouraged to minimise research waste. 73

CONCLUSION

PRO use in DFOT has increased over time, and across a wider variety of settings. However, overall use remains limited. Few trials reported their PRO results and the quality of the PRO reporting was highly variable. Further work is needed to understand key stakeholder attitudes towards PROs in DFOT, and the potential benefits and barriers to their inclusion. Consensus is needed as how to best integrate PRO data into dose‐finding trials. If PROs are to be included in a meaningful way, adaptation of methodological guidelines for protocol content, statistical analysis and reporting will be required to standardise and improve the quality of PRO data in dose‐finding oncology trials.

ETHICS APPROVAL AND CONSENT TO PARTICIPATE

Not applicable.

CONFLICTS OF INTEREST

AM has served on advisory boards and received fees from Merck, FARON, Novartis, Bayer and Janssen, which are all unrelated to this work. CY has received fees from FARON and honorarium from Celgene, which are all unrelated to this work. JLK and ZY declare no conflicts of interest.

AUTHOR CONTRIBUTIONS

JLK, CY – study design, review of ClinicalTrials.gov, analysis of results, manuscript preparation, approval of final manuscript; ZY– analysis of results, manuscript preparation, approval of final manuscript; AM – study design, analysis of results, manuscript preparation, approval of final manuscript.

Supporting information

Table S1

Click here for additional data file.

Text S1

Click here for additional data file.

Data Availability Statement

Data are available on request from the corresponding author.

REFERENCES

1 Atkinson TM , Li Y , Coffey CW , et al. Reliability of adverse symptom event reporting by clinicians . Qual Life Res . 2012 ; 21 ( 7 ): 1159 ‐ 1164 . 21984468 10.1007/s11136-011-0031-4 PMC3633532

2 Basch E , Jia X , Heller G , et al. Adverse symptom event reporting by patients vs clinicians: relationships with clinical outcomes . J Natl Cancer Inst . 2009 ; 101 ( 23 ): 1624 ‐ 1632 . 19920223 10.1093/jnci/djp386 PMC2786917

3 Fromme EK , Eilers KM , Mori M , Hsieh YC , Beer TM . How accurate is clinician reporting of chemotherapy adverse effects? A comparison with patient‐reported symptoms from the Quality‐of‐Life Questionnaire C30 . J Clin Oncol . 2004 ; 22 ( 17 ): 3485 ‐ 3490 . 15337796 10.1200/JCO.2004.03.025

4 Basch E , Iasonos A , McDonough T , et al. Patient versus clinician symptom reporting using the National Cancer Institute Common Terminology Criteria for Adverse Events: results of a questionnaire‐based study . Lancet Oncol . 2006 ; 7 ( 11 ): 903 ‐ 909 . 17081915 10.1016/S1470-2045(06)70910-X

5 Di Maio M , Gallo C , Leighl NB , et al. Symptomatic toxicities experienced during anticancer treatment: agreement between patient and physician reporting in three randomized trials . J Clin Oncol . 2015 ; 33 ( 8 ): 910 ‐ 915 . 25624439 10.1200/JCO.2014.57.9334

6 Harrington JA , Hernandez‐Guerrero TC , Basu B . Early phase clinical trial designs: state of play and adapting for the future . Clin Oncol . 2017 ; 29 ( 12 ): 770 ‐ 777 . 10.1016/j.clon.2017.10.005 29108786

7 Postel‐Vinay S , Gomez‐Roca C , Molife LR , et al. Phase I trials of molecularly targeted agents: should we pay more attention to late toxicities? J Clin Oncol . 2011 ; 29 ( 13 ): 1728 ‐ 1735 . 21444876 10.1200/JCO.2010.31.9236

8 Basch E . Toward patient‐centered drug development in oncology . N Engl J Med . 2013 ; 369 ( 5 ): 397 ‐ 400 . 23822654 10.1056/NEJMp1114649

9 Calvert MJ , O'Connor DJ , Basch EM . Harnessing the patient voice in real‐world evidence: the essential role of patient‐reported outcomes . Nat Rev Drug Discov . 2019 ; 18 ( 10 ): 731 ‐ 732 . 31570837 10.1038/d41573-019-00088-7

10 Kluetz PG , O'Connor DJ , Soltys K . Incorporating the patient experience into regulatory decision making in the USA, Europe, and Canada . Lancet Oncol . 2018 ; 19 ( 5 ): e267 ‐ e274 . 29726391 10.1016/S1470-2045(18)30097-4

11 Food and Drug Administration . Guidance for industry — patient‐reported outcomes measures: use in medical product development to support labeling claims . Silver Spring ; 2009 . 10.1186/1477-7525-4-79 PMC1629006 17034633

12 Cleeland CS , Wang XS , Shi Q , et al. Automated symptom alerts reduce postoperative symptom severity after cancer surgery: a randomized controlled clinical trial . J Clin Oncol . 2011 ; 29 ( 8 ): 994 ‐ 1000 . 21282546 10.1200/JCO.2010.29.8315 PMC3068055

13 Gilbert JE , Howell D , King S , et al. Quality improvement in cancer symptom assessment and control: the Provincial Palliative Care Integration Project (PPCIP) . J Pain Symptom Manage . 2012 ; 43 ( 4 ): 663 ‐ 678 . 22464352 10.1016/j.jpainsymman.2011.04.028

14 Valderas JM , Kotzeva A , Espallargues M , et al. The impact of measuring patient‐reported outcomes in clinical practice: a systematic review of the literature . Qual Life Res . 2008 ; 17 ( 2 ): 179 ‐ 193 . 18175207 10.1007/s11136-007-9295-0

15 Chen J , Ou L , Hollis SJ . A systematic review of the impact of routine collection of patient reported outcome measures on patients, providers and health organisations in an oncologic setting . BMC Health Serv Res . 2013 ; 13 : 211 . 23758898 10.1186/1472-6963-13-211 PMC3700832

16 Kotronoulas G , Kearney N , Maguire R , et al. What is the value of the routine use of patient‐reported outcome measures toward improvement of patient outcomes, processes of care, and health service outcomes in cancer care? A systematic review of controlled trials . J Clin Oncol . 2014 ; 32 ( 14 ): 1480 ‐ 1501 . 24711559 10.1200/JCO.2013.53.5948

17 Detmar SB , Muller MJ , Schornagel JH , Wever LD , Aaronson NK . Health‐related quality‐of‐life assessments and patient‐physician communication: a randomized controlled trial . JAMA . 2002 ; 288 ( 23 ): 3027 ‐ 3034 . 12479768 10.1001/jama.288.23.3027

18 Basch E , Artz D , Dulko D , et al. Patient online self‐reporting of toxicity symptoms during chemotherapy . J Clin Oncol . 2005 ; 23 ( 15 ): 3552 ‐ 3561 . 15908666 10.1200/JCO.2005.04.275

19 Velikova G , Booth L , Smith AB , et al. Measuring quality of life in routine oncology practice improves communication and patient well‐being: a randomized controlled trial . J Clin Oncol . 2004 ; 22 ( 4 ): 714 ‐ 724 . 14966096 10.1200/JCO.2004.06.078

20 Berry DL , Blumenstein BA , Halpenny B , et al. Enhancing patient‐provider communication with the electronic self‐report assessment for cancer: a randomized trial . J Clin Oncol . 2011 ; 29 ( 8 ): 1029 ‐ 1035 . 21282548 10.1200/JCO.2010.30.3909 PMC3068053

21 Nipp RD , Temel JS . Harnessing the power of patient‐reported outcomes in oncology . Clin Cancer Res . 2018 ; 24 : 1777 ‐ 1779 . 29437788 10.1158/1078-0432.CCR-18-0037 PMC5899629

22 Marcus L , Donoghue M , Aungst S , et al. FDA approval summary: entrectinib for the treatment of NTRK gene fusion solid tumors . Clin Cancer Res . 2021 ; 27 : 928 ‐ 932 . 32967940 10.1158/1078-0432.CCR-20-2771

23 Scoggins JF , Patrick DL . The use of patient‐reported outcomes instruments in registered clinical trials: evidence from ClinicalTrials.gov . Contemp Clin Trials . 2009 ; 30 ( 4 ): 289 ‐ 292 . 19275948 10.1016/j.cct.2009.02.005 PMC2683916

24 Fiteni F , Ray IL , Ousmen A , Isambert N , Anota A , Bonnetain F . Health‐related quality of life as an endpoint in oncology phase I trials: a systematic review . BMC Cancer . 2019 ; 19 ( 1 ): 361 . 30991990 10.1186/s12885-019-5579-3 PMC6469065

25 Veitch ZW , Shepshelovich D , Gallagher C , et al. Implementation of PRO‐CTCAE in phase I clinical trials identifies under reporting of adverse events . Ann Oncol . 2019 ; 30 : v718 ‐ v746 .

26 Shepshelovich D , McDonald K , Spreafico A , et al. Feasibility assessment of using the complete patient‐reported outcomes version of the common terminology criteria for adverse events (PRO‐CTCAE) item library . Oncologist . 2019 ; 24 ( 4 ): e146 ‐ e148 . 30728278 10.1634/theoncologist.2018-0332 PMC6459233

27 Veitch ZW , Shepshelovich D , Gallagher C , et al. Underreporting of symptomatic adverse events in phase I clinical trials . J Natl Cancer Inst . 2021 ; 113 : 980 ‐ 988 . 33616650 10.1093/jnci/djab015 PMC8502480

28 Watson GA , Veitch ZWN , Shepshelovich D , et al. Mapping PRO‐CTCAE responses to clinician‐graded adverse events, dose reductions, interruptions, and discontinuations in phase I cancer trials . J Clin Oncol . 2020 ; 38 ( 15 Suppl ): 2014 .32243222

29 Gresham G , Meinert JL , Gresham AG , Meinert CL . Assessment of Trends in the Design, Accrual, and Completion of Trials Registered in ClinicalTrials.gov by Sponsor Type, 2000‐2019 . JAMA Network Open . 2020 ; 3 ( 8 ): e2014682 . 32845329 10.1001/jamanetworkopen.2020.14682 PMC7450351

30 Calvert M , Kyte D , Mercieca‐Bebber R , et al. Guidelines for inclusion of patient‐reported outcomes in clinical trial protocols: the SPIRIT‐PRO extension . JAMA . 2018 ; 319 ( 5 ): 483 ‐ 494 . 29411037 10.1001/jama.2017.21903

31 Laetsch TW , DuBois SG , Mascarenhas L , et al. Larotrectinib for paediatric solid tumours harbouring NTRK gene fusions: phase 1 results from a multicentre, open‐label, phase 1/2 study . Lancet Oncol . 2018 ; 19 ( 5 ): 705 ‐ 714 . 29606586 10.1016/S1470-2045(18)30119-0 PMC5949072

32 Taylor SE , Li R , Petschauer JS , et al. Phase I study of intravenous (IV) docetaxel and intraperitoneal (IP) oxaliplatin in recurrent ovarian and fallopian tube cancer . Gynecol Oncol . 2015 ; 138 ( 3 ): 548 ‐ 553 . 26111788 10.1016/j.ygyno.2015.06.026

33 Taylor SE , Petschauer JS , Donovan H , et al. Phase I study of intravenous oxaliplatin and intraperitoneal docetaxel in recurrent ovarian cancer . Int Jo Gynecol Cancer . 2019 ; 29 ( 1 ): 147 ‐ 152 . 10.1136/ijgc-2018-000055 30640697

34 Wyatt GK , Sikorskii A , Safikhani A , McVary KT , Herman J . Saw palmetto for symptom management during radiation therapy for prostate cancer . J Pain Symptom Manage . 2016 ; 51 ( 6 ): 1046 ‐ 1054 . 26891611 10.1016/j.jpainsymman.2015.12.315

35 van Zandwijk N , Pavlakis N , Kao SC , et al. Safety and activity of microRNA‐loaded minicells in patients with recurrent malignant pleural mesothelioma: a first‐in‐man, phase 1, open‐label, dose‐escalation study . Lancet Oncol . 2017 ; 18 ( 10 ): 1386 ‐ 1396 . 28870611 10.1016/S1470-2045(17)30621-6

36 Haddad T , Qin R , Lupu R , et al. A phase I study of cilengitide and paclitaxel in patients with advanced solid tumors . Cancer Chemother Pharmacol . 2017 ; 79 ( 6 ): 1221 ‐ 1227 . 28477227 10.1007/s00280-017-3322-9 PMC9594487

37 Sampath S , Frankel P , Vecchio BD , et al. Stereotactic body radiation therapy to the prostate bed: results of a phase 1 dose‐escalation trial . Int J Radiat Oncol Biol Phys . 2020 ; 106 ( 3 ): 537 ‐ 545 . 31733323 10.1016/j.ijrobp.2019.11.005

38 Reiss KA , Herman JM , Zahurak M , et al. A Phase I study of veliparib (ABT‐888) in combination with low‐dose fractionated whole abdominal radiation therapy in patients with advanced solid malignancies and peritoneal carcinomatosis . Clin Cancer Res . 2015 ; 21 ( 1 ): 68 ‐ 76 . 25355929 10.1158/1078-0432.CCR-14-1552 PMC4286495

39 Dinkic C , Eichbaum M , Schmidt M , et al. Pazopanib (GW786034) and cyclophosphamide in patients with platinum‐resistant, recurrent, pre‐treated ovarian cancer ‐ Results of the PACOVAR‐trial . Gynecol Oncol . 2017 ; 146 ( 2 ): 279 ‐ 284 . 28528917 10.1016/j.ygyno.2017.05.013

40 Aguilar LK , Shirley LA , Chung VM , et al. Gene‐mediated cytotoxic immunotherapy as adjuvant to surgery or chemoradiation for pancreatic adenocarcinoma . Cancer Immunol Immunother . 2015 ; 64 ( 6 ): 727 ‐ 736 . 25795132 10.1007/s00262-015-1679-3 PMC11029723

41 Subbiah V , Grilley‐Olson JE , Combest AJ , et al. Phase Ib/II trial of NC‐6004 (Nanoparticle Cisplatin) plus gemcitabine in patients with advanced solid tumors . Clin Cancer Res . 2018 ; 24 ( 1 ): 43 ‐ 51 . 29030354 10.1158/1078-0432.CCR-17-1114

42 Crew KD , Brown P , Greenlee H , et al. Phase IB randomized, double‐blinded, placebo‐controlled, dose escalation study of polyphenon E in women with hormone receptor‐negative breast cancer . Cancer Prev Res . 2012 ; 5 ( 9 ): 1144 ‐ 1154 . 10.1158/1940-6207.CAPR-12-0117 PMC3816771 22827973

43 Kumar SK , Buadi FK , LaPlant B , et al. Phase 1/2 trial of ixazomib, cyclophosphamide and dexamethasone in patients with previously untreated symptomatic multiple myeloma . Blood Cancer J . 2018 ; 8 ( 8 ): 70 . 30061664 10.1038/s41408-018-0106-3 PMC6066484

44 Verstovsek S , Kantarjian H , Mesa RA , et al. Safety and efficacy of INCB018424, a JAK1 and JAK2 inhibitor, myelofibrosis . N Engl J Med . 2010 ; 363 ( 12 ): 1117 ‐ 1127 . 20843246 10.1056/NEJMoa1002028 PMC5187954

45 Verstovsek S , Talpaz M , Ritchie E , et al. A phase I, open‐label, dose‐escalation, multicenter study of the JAK2 inhibitor NS‐018 in patients with myelofibrosis . Leukemia . 2017 ; 31 ( 2 ): 393 ‐ 402 . 27479177 10.1038/leu.2016.215 PMC5292677

46 Boulin M , Hillon P , Cercueil JP , et al. Idarubicin‐loaded beads for chemoembolisation of hepatocellular carcinoma: results of the IDASPHERE phase I trial . Aliment Pharmacol Ther . 2014 ; 39 ( 11 ): 1301 ‐ 1313 . 24738629 10.1111/apt.12746

47 Siegel AB , Narayan R , Rodriguez R , et al. A phase I dose‐finding study of silybin phosphatidylcholine (milk thistle) in patients with advanced hepatocellular carcinoma . Integr Cancer Ther . 2014 ; 13 ( 1 ): 46 ‐ 53 . 23757319 10.1177/1534735413490798 PMC3866213

48 Martin‐Broto J , Hindi N , Lopez‐Pousa A , et al. Assessment of safety and efficacy of combined trabectedin and low‐dose radiotherapy for patients with metastatic soft‐tissue sarcomas: a nonrandomized phase 1/2 clinical trial . JAMA Oncol . 2020 ; 6 ( 4 ): 535 ‐ 541 . 32077895 10.1001/jamaoncol.2019.6584 PMC7042948

49 Watanabe S , Yoshioka H , Sakai H , et al. Necitumumab plus gemcitabine and cisplatin versus gemcitabine and cisplatin alone as first‐line treatment for stage IV squamous non‐small cell lung cancer: A phase 1b and randomized, open‐label, multicenter, phase 2 trial in Japan . Lung Cancer . 2019 ; 129 : 55 ‐ 62 . 30797492 10.1016/j.lungcan.2019.01.005

50 Wong KM , Capasso A , Eckhardt SG . The changing landscape of phase I trials in oncology . Nat Rev Clin Oncol . 2016 ; 13 ( 2 ): 106 ‐ 117 . 26552953 10.1038/nrclinonc.2015.194

51 Le Tourneau C , Gan HK , Razak AR , Paoletti X . Efficiency of new dose escalation designs in dose‐finding phase I trials of molecularly targeted agents . PLoS One . 2012 ; 7 ( 12 ): e51039 . 23251419 10.1371/journal.pone.0051039 PMC3521009

52 Manji A , Brana I , Amir E , et al. Evolution of clinical trial design in early drug development: systematic review of expansion cohort use in single‐agent phase I cancer trials . J Clin Oncol . 2013 ; 31 ( 33 ): 4260 ‐ 4267 . 24127441 10.1200/JCO.2012.47.4957

53 Riviere MK , Le Tourneau C , Paoletti X , Dubois F , Zohar S . Designs of drug‐combination phase I trials in oncology: a systematic review of the literature . Ann Oncol . 2015 ; 26 ( 4 ): 669 ‐ 674 . 25403591 10.1093/annonc/mdu516

54 Basch E , Yap C . Patient‐reported outcomes for tolerability assessment in phase I cancer clinical trials . J Natl Cancer Inst . 2021 ; 113 ( 8 ): 943 ‐ 944 . 33616647 10.1093/jnci/djab017 PMC8328974

55 Trask PC , Dueck AC , Piault E , Campbell A . Patient‐reported outcomes version of the common terminology criteria for adverse events: methods for item selection in industry‐sponsored oncology clinical trials . Clin Trials . 2018 ; 15 ( 6 ): 616 ‐ 623 . 30230365 10.1177/1740774518799985

56 Reeve BB , Mitchell SA , Dueck AC , et al. Recommended patient‐reported core set of symptoms to measure in adult cancer treatment trials . J Natl Cancer Inst . 2014 ; 106 ( 7 ). 10.1093/jnci/dju129 PMC4110472 25006191

57 Sodergren SC , Wheelwright SJ , Fitzsimmons D , et al. Developing symptom lists for people with cancer treated with targeted therapies . Target. Oncol . 2021 ; 16 ( 1 ): 95 ‐ 107 . 33165751 10.1007/s11523-020-00769-z

58 Chung AE , Shoenbill K , Mitchell SA , et al. Patient free text reporting of symptomatic adverse events in cancer clinical research using the National Cancer Institute’s Patient‐Reported Outcomes version of the Common Terminology Criteria for Adverse Events (PRO‐CTCAE) . J Am Med Inform Assoc . 2019 ; 26 ( 4 ): 276 ‐ 285 . 30840079 10.1093/jamia/ocy169 PMC6402312

59 Schandelmaier S , Conen K , von Elm E , et al. Planning and reporting of quality‐of‐life outcomes in cancer trials . Ann Oncol . 2015 ; 26 ( 9 ): 1966 ‐ 1973 . 26133966 10.1093/annonc/mdv283 PMC4551161

60 Weingärtner V , Dargatz N , Weber C , et al. Patient reported outcomes in randomized controlled cancer trials in advanced disease: a structured literature review . Expert Rev Clin Pharmacol . 2016 ; 9 ( 6 ): 821 ‐ 829 . 26959869 10.1586/17512433.2016.1164595

61 Kyte D , Retzer A , Ahmed K , et al. Systematic evaluation of patient‐reported outcome protocol content and reporting in cancer trials . J Natl Cancer Inst . 2019 ; 111 ( 11 ): 1170 ‐ 1178 . 30959516 10.1093/jnci/djz038 PMC6855977

62 Bylicki O , Gan HK , Joly F , Maillet D , You B , Péron J . Poor patient‐reported outcomes reporting according to CONSORT guidelines in randomized clinical trials evaluating systemic cancer therapy . Ann Oncol . 2015 ; 26 ( 1 ): 231 ‐ 237 . 25355720 10.1093/annonc/mdu489

63 Smith AB , Cocks K , Parry D , Taylor M . Reporting of health‐related quality of life (HRQOL) data in oncology trials: a comparison of the European Organization for Research and Treatment of Cancer Quality of Life (EORTC QLQ‐C30) and the Functional Assessment of Cancer Therapy‐General (FACT‐G) . Qual Life Res . 2014 ; 23 ( 3 ): 971 ‐ 976 . 24097080 10.1007/s11136-013-0534-2

64 Efficace F , Fayers P , Pusic A , et al. Quality of patient‐reported outcome reporting across cancer randomized controlled trials according to the CONSORT patient‐reported outcome extension: a pooled analysis of 557 trials . Cancer . 2015 ; 121 ( 18 ): 3335 ‐ 3342 . 26079197 10.1002/cncr.29489 PMC4755292

65 Laird NM , Ware JH . Random‐effects models for longitudinal data . Biometrics . 1982 ; 38 ( 4 ): 963 ‐ 974 . 7168798

66 Liang K‐Y , Zeger SL . Longitudinal data analysis using generalized linear models . Biometrika . 1986 ; 73 ( 1 ): 13 ‐ 22 .

67 DeVito NJ , Goldacre B . Evaluation of compliance with legal requirements under the FDA amendments act of 2007 for timely registration of clinical trials, data verification, delayed reporting, and trial document submission . JAMA Intern Med . 2021 ; 181 ( 8 ): 1128 . 34028509 10.1001/jamainternmed.2021.2036 PMC8145152

68 Shepshelovich D , Goldvaser H , Wang L , Abdul Razak AR . Comparison of published and unpublished phase I clinical cancer trials: an analysis of the CliniclTrials.gov database . Invest New Drugs . 2018 ; 36 ( 5 ): 933 ‐ 938 . 29234946 10.1007/s10637-017-0549-6

69 Clinical Trials Transformation Initiative . Points to consider for statistical analysis using the database for aggregate analysis of ClinicalTrials.gov . 2013 . https://www.ctti‐clinicaltrials.org/file/aact2013statisticalpointstoconsider122713docx [Accessed 13 May, 2020].

70 Kennedy F , Shearsmith L , Ayres M , et al. Online monitoring of patient self‐reported adverse events in early phase clinical trials: Views from patients, clinicians, and trial staff . Clin Trials . 2021 ; 18 ( 2 ): 168 ‐ 179 . 33231103 10.1177/1740774520972125 PMC8010887

71 Calvert M , Blazeby J , Altman DG , et al. Reporting of patient‐reported outcomes in randomized trials: the CONSORT PRO extension . JAMA . 2013 ; 309 ( 8 ): 814 ‐ 822 . 23443445 10.1001/jama.2013.879

72 Coens C , Pe M , Dueck AC , et al. International standards for the analysis of quality‐of‐life and patient‐reported outcome endpoints in cancer randomised controlled trials: recommendations of the SISAQOL Consortium . Lancet Oncol . 2020 ; 21 ( 2 ): e83 ‐ e96 . 32007209 10.1016/S1470-2045(19)30790-9

73 Cella D . In our patient‐centered era, it is time we gave patient‐reported outcomes their due . Cancer . 2020 ; 126 ( 11 ): 2592 ‐ 2593 . 32237234 10.1002/cncr.32763 PMC7220818