Abstract
Background
Genotype-matched trial participation often requires referrals from institutions offering comprehensive genomic profiling (CGP) testing to those conducting these trials, entailing potential financial burdens. The impact of travel costs on trial participation remains unclear.
Patients and methods
This retrospective cohort study (2020-2022) evaluated 1127 patients with metastatic solid tumors referred to a single cancer center for trial participation following CGP testing. Primary and secondary outcomes were participation in genotype-matched and genotype-nonmatched trials, respectively. Multivariable logistic regression assessed the association between travel costs and trial participation.
Results
Of the 1127 patients, 127 (11%) participated in genotype-matched trials and 114 (10%) in genotype-nonmatched trials. The median travel cost (round-trip public transportation) was $17 (range $3-$982). Multivariable regression showed that patients with travel costs ≥$100 were significantly less likely to participate in genotype-matched trials than those with travel costs <$100 [7% versus 13%; odds ratio (OR) = 0.51, 95% confidence interval (CI) 0.30-0.88]. Genotype-matched trial participation likelihood decreased as travel costs increased from <$100 to $100-200 and ≥$200 (13% versus 9% versus 6%; OR = 0.70, 95% CI 0.28-1.52; OR = 0.46, 95% CI 0.22-0.85). Travel costs did not affect genotype-nonmatched trial participation. Genotype-matched trial participants had a shorter median time from initial clinic appointment to trial participation than genotype-nonmatched trial participants (21 versus 31 days, P = 0.006).
Conclusion
Increased travel costs were associated with a decreased likelihood of genotype-matched trial participation after CGP testing, highlighting the need for reformed clinical trial platforms to alleviate financial travel burdens.
Key words: phase I, DCT, financial toxicity, travel burden, decentralized clinical trials, travel time
Graphical abstract
Highlights
-
•
Genotype-matched trials need a referral to cancer centers, raising travel costs.
-
•
Travel costs link to lower genotype-matched trial enrollment, but not nonmatched.
-
•
Genotype-matched trials have a shorter time from the first visit to participation.
-
•
Financial travel burdens may create inequities in precision oncology access.
-
•
Decentralizing trials and financial assistance could mitigate access disparities.
Introduction
The advent of next-generation sequencing and genomic-specific therapeutics have increased the use of comprehensive genomic profiling (CGP) to direct patients toward genotype-matched therapies. Over the past two decades, the genomic alterations suitable for genotype-matched trials have consistently grown.1 Participation in genotype-matched trials offers patients the opportunity to receive high-quality care and novel therapies. This progress highlights that inequities in clinical trial access present a pressing ethical challenge.2 The centralization of clinical trials at major academic hospitals imposes challenges on community hospital oncologists, 42% of whom have reported difficulty in finding appropriate clinical trials within a reasonable travel distance for their patients.3 Substantial disparities exist in trial access and participation, especially among patients with cancer residing in rural areas.3, 4, 5
In a survey conducted by the American Society of Clinical Oncology (ASCO) among rural and urban patients who did not participate in a clinical trial, 14% raised concerns regarding the need for long travel distances to trail sites as a reason for their nonparticipation.3 We have recently demonstrated that increased travel time and distance are associated with a decreased likelihood of genotype-matched trial participation, suggesting that travel burden contributes to inequities in precision oncology.5 Clinical trials impose a significant financial burden on patients.6,7 Previous research showed that half of the patients expressed concerns regarding the cost of clinical trial participation7, 8, 9 and that travel costs could hinder trial participation as a financial burden.10, 11, 12 However, the relationship between travel costs and participation in genotype-matched trials remains nebulous.
The overall patient participation rate in cancer clinical trials is ∼5%.13 CGP testing provides patients with increased access to trials, yet only 10%-20% of those undergoing CGP testing receive genotype-matched therapies.14, 15, 16, 17, 18 Travel burdens, including travel time, distance, and costs, can create physical, psychological, and financial barriers,5, 6, 7 potentially leading to prolonged time from the initial outpatient appointment to trial participation and disease progression, thereby hindering subsequent trial participation. However, data on the relationship between travel burdens (cost, distance, time) and the time from the initial outpatient appointment to trial participation remain lacking.
Herein, we investigated the association between travel costs and participation in genotype-matched trials and genotype-nonmatched trials following CGP testing and referral to the National Cancer Center Hospital (NCCH), Tokyo. In addition, the relationship between travel burden (travel time, distance, and costs) and the time from the initial outpatient appointment to trial participation was investigated in the trial matching by genotype (genotype-matched versus genotype-nonmatched trial). This research of travel costs in relation to participation in clinical trials could help guide future strategies to make precision oncology accessible to all patients.
Methods
Study design
Between June 2020 and June 2022, we reviewed 1127 patients referred to the NCCH by physicians from other institutions following CGP testing and discussions with the regional molecular tumor board (MTB). Most patients had advanced or metastatic solid tumors and completed or nearly completed standard therapies, barring a few exceptions for those without standard therapy options. Clinical data were sourced from electronic medical records, with data acquisition completed on 30 October 2022. The following data were collected: demographics, previous medical history, molecular profiling results, laboratory tests conducted around the date of the initial visit, and reasons for participation or nonparticipation after screening in clinical trials. Nonparticipation in clinical trials is often attributed to multiple reasons, with the primary one being determined as the most critical clinical factor influencing participation.5 Data were collected by two medical oncologists (YU and TK) and a trained clinical research coordinator (R. Tsunoda). The NCCH Institutional Review Board approved this research (approval number 2014-148).
Comprehensive genomic profiling testing
CGP testing was conducted using hybrid capture-based targeted DNA sequencing with FoundationOne CDx, the OncoGuide NCC Oncopanel System, or in-house next-generation sequencing testing or in blood-derived cell-free DNA with FoundationOne Liquid CDx or Guardant360 CDx.19, 20, 21, 22 After deliberations among regional MTB, including oncologists, pediatric oncologists, pathologists, bioinformaticians, genome researchers, and genetic counselors, all patients were referred to the NCCH.16,23 Reports from inspection companies, alongside reports featuring clinical annotations and details regarding genotype-matched clinical trials from the Center for Cancer Genomics and Advanced Therapeutics national database, were employed by MTB in their evaluations.24, 25, 26
Outcomes
The primary outcome of this research was patient participation in genotype-matched trials, defined as trials involving patient enrollment based on the results of CGP testing and the recommendations of regional MTB. The validity of genotype-matched trials was reviewed by multiple investigators in this research. The specific genomic alterations and corresponding therapies used in these genotype-matched trials are listed in Supplementary Table S1, available at https://doi.org/10.1016/j.esmorw.2025.100114. The secondary outcome was patient participation in genotype-nonmatched trials.
Definition of variables
Travel costs, distance, or time were calculated using Yahoo! Maps (https://map.yahoo.co.jp/) for round-trip public transportation from the patient’s residence to the NCCH at 12:00 am on a Monday. Data on travel costs were dichotomized as <$100 versus ≥$100 (1$ = 120 JPY).
Based on prior studies and expert agreement, 23 specific covariates were selected, namely, age, sex, Eastern Cooperative Oncology Group performance status, body mass index (BMI), tumor type (common versus noncommon cancers), number of lines of prior therapies, number of metastatic sites, liver metastases, brain metastases, pleural effusion or ascites, suitability for biopsy (whether the tumor can be subjected to biopsy), neutrophils, hemoglobin, platelets, albumin, creatinine, total bilirubin, lactate dehydrogenase, aspartate aminotransferase, C-reactive protein, place of residence (suburban/rural versus urban), types of referring hospitals (community hospital versus university hospital or cancer center), and household income.8,27,28 The term ‘common cancers’ has been used to describe the top 10 cancer types leading in mortality as identified by the World Health Organization, which are cancers of the lung, colorectum, liver, stomach, breast, esophagus, pancreas, prostate, cervix, and endometrium.29 Biopsy-eligible lesions are identified as metastases to the lungs and liver, tumors in the pelvic area, lymph nodes, and superficial sites on the skin, head, neck, or chest wall that are accessible for biopsy through percutaneous methods, computed tomography guidance, bronchoscopy, or endoscopy. Such imaging was reviewed by our team of medical oncologists and radiologists specializing in diagnostics and interventions to evaluate lesion accessibility. In addition, cities with populations exceeding 1 million, such as Sapporo, Sendai, Saitama, Tokyo, Yokohama, Kawasaki, Chiba, Nagoya, Kyoto, Osaka, Kobe, Hiroshima, and Fukuoka, were categorized as urban areas. The median household income data were derived from the 2020 census reports based on zip codes.30
Statistical analysis
Baseline patient characteristics
Patient characteristics were summarized using descriptive statistical methods. The chi-square (χ2) test was used for categorical variables, whereas the Mann–Whitney test was used for continuous variables. Numeric variables were divided into two groups using established standards values, when available; for variables without such references, the division was guided by prior research or the consensus among experts.16,28,31,32
Analysis of the association between travel costs and both genotype-matched and genotype-nonmatched trial participation
An analysis of the association between travel costs and genotype-matched and genotype-nonmatched trial participation was conducted using multivariable logistic regression. The multivariable model included all independent variables associated with each clinical trial participation with a P value of <0.20. The variance inflation factor was used to evaluate collinearity within the regression model. Multiple imputation was employed for variables with missing data—BMI, neutrophils, hemoglobin, platelets, albumin, lactate dehydrogenase, creatinine, total bilirubin, aspartate aminotransferase, and C-reactive protein—assuming the data were missing at random.33 In the sensitivity analysis, travel costs were categorized into <$100, $100-$200, or ≥$200. Moreover, a complete case analysis was carried out among participants with reported known BMI and laboratory values.
Comparison of travel costs with travel distance and travel time
Spearman’s correlation coefficients were determined to examine the correlation between travel costs and travel distance/time. Spearman’s correlation values of 0.30-0.50 were considered low; 0.50-0.70, moderate; 0.70-0.90, high.34
Comparison of trial matching by genotype and travel burden, and the time from the initial outpatient appointment to trial participation
The Mann–Whitney test was used to examine the association between the time from the initial outpatient appointment to trial participation and trial matching by genotype (genotype-matched versus genotype-nonmatched trial). Spearman’s correlation coefficients were calculated to assess the correlation between travel burden (time, distance, and costs) and the time from the initial outpatient appointment to trial participation in the trial matching by genotype. All statistical analyses were carried out using R statistical software, version 4.2.3 (R Foundation, Vienna, Austria). Significance was set at a two-sided P value of < 0.05.
Results
Patient characteristics
Among the 1127 patients assessed, 127 (11%) participated in genotype-matched trials and 114 (10%) in genotype-nonmatched trials (Figure 1). Patient characteristics are listed in Table 1. The overall median travel cost was $17 [interquartile range (IQR) $9-$80]. The median travel cost for participants in genotype-matched trials was $15 (IQR $8-$50), and $20 (IQR $10-$64) for those in genotype-nonmatched trials. Patient details in both trial types are presented in Supplementary Table S2, available at https://doi.org/10.1016/j.esmorw.2025.100114. Of the genotype-matched trial participants, 82 (65%) participated in phase I trials and 45 (35%) in phase II/III trials; most of these trials were industry-sponsored (92%) and targeted drug monotherapy (69%). In the genotype-nonmatched trial participants, 102 (90%) participated in phase I trials and 12 (10%) in phase II/III trials. All genotype-nonmatched trials were industry-sponsored (100%), with immunotherapy (47%) and targeted drug monotherapy (44%) being the most common types of therapy.
Figure 1.
Consolidated Standards of Reporting Trials (CONSORT) diagram illustrating patient flow from referral to genotype-matched and genotype-nonmatched trial participation. NCCH, National Cancer Center Hospital.
Table 1.
Characteristics of patients stratified by enrollment in genotype-matched trials or genotype-nonmatched trials (n = 1127)
| Demographics | All patients (n = 1127) | Enrolled in genotype-matched trials (n = 127) | Not enrolled in genotype-matched trials (n = 1000) | P | Enrolled in genotype-nonmatched trials (n = 114) | Not enrolled in genotype-nonmatched trials (n = 1013) | P |
|---|---|---|---|---|---|---|---|
| Age, median (range), years | 62 (16-85) | 59 (20-80) | 62 (16-85) | 0.013 | 60 (38-79) | 62 (16-85) | 0.083 |
| Sex, n (%) | 0.75 | 0.27 | |||||
| Men | 543 (48) | 59 (47) | 484 (48) | 61 (53) | 482 (48) | ||
| Women | 584 (52) | 68 (53) | 516 (52) | 53 (47) | 531 (52) | ||
| Performance status, n (%) | <0.001 | <0.001 | |||||
| 0 | 434 (39) | 72 (57) | 362 (36) | 74 (65) | 360 (36) | ||
| 1 | 607 (54) | 54 (43) | 553 (55) | 40 (35) | 567 (56) | ||
| ≥2 | 86 (8) | 1 (1) | 85 (9) | 0 (0) | 86 (9) | ||
| Body mass indexa, n (%) | 0.99 | <0.001 | |||||
| <18.5 | 161 (14) | 18 (14) | 143 (14) | 9 (8) | 152 (15) | ||
| 18.5-25 | 775 (69) | 88 (69) | 687 (69) | 72 (63) | 703 (70) | ||
| ≥25 | 189 (17) | 21 (17) | 168 (17) | 33 (29) | 156 (15) | ||
| Cancer type | 0.36 | 0.046 | |||||
| Gastrointestinal | 335 (30) | 44 (35) | 291 (29) | 42 (37) | 293 (29) | ||
| Hepatobiliary | 314 (28) | 30 (24) | 284 (28) | 26 (23) | 288 (28) | ||
| Gynecology | 179 (16) | 22 (17) | 157 (16) | 18 (16) | 161 (16) | ||
| Breast | 61 (5) | 4 (3) | 57 (6) | 5 (4) | 56 (6) | ||
| Lung | 56 (5) | 9 (7) | 47 (5) | 9 (8) | 47 (5) | ||
| Sarcoma | 43 (4) | 8 (6) | 35 (4) | 3 (3) | 40 (4) | ||
| Genitourinary | 42 (4) | 4 (3) | 38 (4) | 6 (5) | 36 (4) | ||
| Head and neck | 23 (2) | 3 (2) | 20 (2) | 1 (0.9) | 22 (2) | ||
| Skin, melanoma | 14 (1) | 0 (0) | 14 (1) | 1 (0.9) | 13 (1) | ||
| Central nervous system | 14 (1) | 1 (1) | 13 (1) | 0 (0) | 14 (1) | ||
| Carcinoma of unknown primary | 10 (1) | 1 (1) | 9 (1) | 0 (0) | 10 (1) | ||
| Otherb | 36 (3) | 1 (1) | 35 (4) | 3 (3) | 33 (3) | ||
| Number of lines of prior therapies, n (%) | 0.93 | 0.005 | |||||
| Median (range) | 3 (0-12) | 3 (0-12) | 3 (1-8) | 3 (0-12) | 3 (0-11) | ||
| <2 | 176 (16) | 19 (15) | 157 (16) | 7 (6) | 169 (17) | ||
| ≥2 | 951 (84) | 108 (85) | 843 (84) | 107 (94) | 844 (83) | ||
| Number of metastatic sites, n (%) | 0.26 | 0.064 | |||||
| Median (range) | 2 (0-7) | 2 (1-5) | 2 (0-7) | 2 (0-5) | 2 (0-7) | ||
| <2 | 354 (31) | 46 (36) | 308 (31) | 45 (40) | 309 (31) | ||
| ≥2 | 773 (69) | 81 (64) | 692 (69) | 69 (60) | 704 (69) | ||
| Liver metastases, n (%) | 0.84 | 0.043 | |||||
| No | 616 (55) | 71 (55) | 545 (54) | 73 (64) | 543 (54) | ||
| Yes | 511 (45) | 56 (45) | 455 (46) | 41 (36) | 470 (46) | ||
| Brain metastases, n (%) | 0.27 | 0.40 | |||||
| No | 1063 (94) | 123 (97) | 940 (94) | 110 (96) | 953 (94) | ||
| Yes | 64 (6) | 4 (3) | 60 (6) | 4 (4) | 60 (6) | ||
| Pleural effusions or ascites, n (%) | 0.003 | 0.002 | |||||
| No | 963 (85) | 120 (95) | 843 (84) | 109 (96) | 854 (84) | ||
| Yes | 164 (15) | 7 (5) | 157 (16) | 5 (4) | 159 (16) | ||
| Suitable for biopsy, n (%) | 0.010 | 0.001 | |||||
| No | 307 (27) | 22 (17) | 285 (28) | 16 (14) | 291 (29) | ||
| Yes | 820 (73) | 105 (83) | 715 (72) | 98 (86) | 722 (71) | ||
| Neutrophils (/μl)c, n (%) | 0.15 | 0.040 | |||||
| <1500 | 59 (5) | 3 (2) | 56 (6) | 1 (1) | 58 (6) | ||
| ≥1500 | 1021 (91) | 124 (98) | 897 (90) | 113 (99) | 908 (90) | ||
| Hemoglobin (g/dl)d, n (%) | 0.013 | 0.004 | |||||
| <9 | 92 (8) | 3 (2) | 89 (9) | 1 (1) | 91 (9) | ||
| ≥9 | 988 (88) | 124 (98) | 864 (86) | 113 (99) | 875 (86) | ||
| Platelets (×103/μl)e, n (%) | 0.20 | 0.052 | |||||
| <100 | 55 (5) | 3 (2) | 52 (5) | 1 (1) | 54 (5) | ||
| ≥100 | 1025 (91) | 124 (98) | 901 (90) | 113 (99) | 912 (90) | ||
| Albumin (g/dl)f, n (%) | <0.001 | <0.001 | |||||
| <3.5 | 257 (23) | 11 (9) | 246 (25) | 8 (7) | 249 (25) | ||
| ≥3.5 | 818 (73) | 116 (91) | 702 (70) | 106 (93) | 712 (70) | ||
| Lactate dehydrogenase (U/dl)g, n (%) | 0.020 | 0.089 | |||||
| <2.5 × upper normal limit | 1000 (89) | 125 (98) | 875 (88) | 111 (97) | 889 (88) | ||
| ≥2.5 × upper normal limit | 74 (7) | 2 (2) | 72 (7) | 3 (3) | 71 (7) | ||
| Creatinine (mg/dl)h, n (%) | 0.76 | 0.33 | |||||
| <1.5 | 1063 (94) | 126 (99) | 937 (94) | 114 (100) | 949 (94) | ||
| ≥1.5 | 16 (1) | 1 (1) | 15 (2) | 0 (0) | 16 (2) | ||
| Total bilirubin (mg/dl)i, n (%) | >0.99 | 0.62 | |||||
| <1.5 × upper normal limit | 1059 (94) | 126 (99) | 933 (93) | 114 (100) | 945 (93) | ||
| ≥1.5 × upper normal limit | 9 (1) | 1 (1) | 8 (1) | 0 (0) | 9 (1) | ||
| Aspartate aminotransferase (U/l)j, n (%) | 0.21 | 0.093 | |||||
| <3.0 × upper normal limit | 1047 (93) | 126 (99) | 921 (92) | 114 (100) | 933 (92) | ||
| ≥3.0 × upper normal limit | 32 (3) | 1 (1) | 31 (3) | 0 (0) | 32 (3) | ||
| C-reactive protein (mg/dl)k, n (%) | 0.002 | 0.002 | |||||
| <5 | 931 (83) | 123 (97) | 808 (81) | 112 (98) | 819 (81) | ||
| ≥5 | 111 (10) | 3 (2) | 108 (11) | 0 (0) | 111 (11) | ||
| Place of residence, n (%) | 0.93 | >0.99 | |||||
| Suburban/Rural | 844 (75) | 96 (74) | 748 (81) | 85 (75) | 759 (75) | ||
| Urban | 283 (25) | 31 (24) | 252 (25) | 29 (25) | 254 (25) | ||
| Referring hospital, n (%) | 0.38 | 0.13 | |||||
| Community hospital | 585 (52) | 61 (48) | 524 (52) | 58 (51) | 527 (52) | ||
| University hospital | 310 (28) | 34 (27) | 276 (28) | 25 (22) | 285 (28) | ||
| Cancer center | 232 (21) | 32 (25) | 200 (20) | 31 (27) | 201 (20) | ||
| Median income (by zip code), n (%) | 0.66 | 0.076 | |||||
| <$25 000 | 124 (11) | 12 (10) | 112 (11) | 14 (12) | 110 (11) | ||
| ≥$25 000 | 1003 (89) | 115 (90) | 888 (89) | 100 (88) | 903 (89) | ||
| Travel time, n (%) | 0.014 | 0.60 | |||||
| Median (interquartile range) | 55 (35-110) | 55 (35-90) | 60 (40-112) | 55 (40-100) | 55 (35-110) | ||
| <120 min | 851 (76) | 108 (85) | 743 (74) | 88 (77) | 763 (75) | ||
| ≥120 min | 276 (24) | 19 (15) | 257 (26) | 26 (23) | 250 (25) | ||
| Travel distance, n (%) | 0.025 | 0.54 | |||||
| Median (interquartile range) | 38 (21-107) | 39 (18-81) | 38 (21-116) | 40 (24-97) | 38 (20-111) | ||
| <100 km | 817 (72) | 101 (80) | 716 (72) | 86 (75) | 731 (72) | ||
| ≥100 km | 310 (28) | 26 (20) | 284 (28) | 28 (25) | 282 (28) | ||
| Travel costs, n (%) | 0.062 | 0.42 | |||||
| Median (interquartile range) | 17 (9-80) | 15 (8-50) | 17 (10-87) | 20 (10-64) | 16 (9-80) | ||
| <$100 | 870 (77) | 109 (86) | 761 (76) | 90 (79) | 780 (77) | ||
| ≥$100 | 257 (23) | 18 (14) | 239 (24) | 24 (21) | 233 (23) |
Missing data for body mass index: n = 2.
Other: neuroendocrine tumor (n = 13), thyroid cancer (n = 8), germ cell tumor (n = 4), thymic cancer (n = 2), mesothelioma (n = 2), adrenal cortical cancer (n = 2), malignant peripheral nerve sheath tumor (n = 1), desmoid tumor (n = 1), Bowen’s disease (n = 1), epithelioid hemangioendothelioma (n = 1), and penile cancer (n = 1).
Missing data for neutrophils: n = 47.
Missing data for hemoglobin: n = 47.
Missing data for platelets: n = 47.
Missing data for albumin: n = 52.
Missing data for lactate dehydrogenase: n = 53.
Missing data for creatinine: n = 48.
Missing data for total bilirubin: n = 59.
Missing data for aspartate aminotransferase: n = 48.
Missing data for C-reactive protein: n = 85.
Association between travel costs and genotype-matched trial participation
After adjusting for covariates, patients with travel costs ≥$100 were significantly less likely to participate in genotype-matched trials than those with travel costs <$100 [7% versus 13%; odds ratio (OR) = 0.51, 95% confidence interval (CI) 0.30-0.88; P = 0.020; Table 2]. The likelihood of genotype-matched trial participation decreased as travel costs increased from <$100 to $100-200 and ≥$200 (13% versus 9% versus 6%, respectively; OR = 0.70, 95% CI 0.28-1.52; OR = 0.46, 95% CI 0.22-0.85; Figure 2A-C). In the sensitivity analysis, the ORs for the complete-case analyses were consistent with those of the primary analysis.
Table 2.
Univariable and multivariable analyses of factors associated with genotype-matched trial and genotype-nonmatched trial participation by travel costs (n = 1127)
| Genotype-matched trial participation | Univariable analysis |
Multivariable analysis |
||||
|---|---|---|---|---|---|---|
| Odds ratio | 95% Confidence interval | P | Odds ratio | 95% Confidence interval | P | |
| Travel costs ≥$100 (versus <$100) | 0.53 | 0.30-0.86 | 0.015 | 0.51 | 0.30-0.88 | 0.020 |
| Age ≥65 years (versus <65 years) | 0.69 | 0.47-0.99 | 0.047 | 0.64 | 0.43-0.94 | 0.024 |
| Sex, women (versus men) | 1.04 | 0.72-1.51 | 0.82 | NA | NA | NA |
| Performance status ≥1 (versus 0) | 0.45 | 0.31-0.65 | <0.001 | 0.57 | 0.39-0.85 | 0.005 |
| Body mass index <18.5 (versus ≥18.5) | 0.87 | 0.52-1.40 | 0.59 | NA | NA | NA |
| Tumor type, noncommon cancers (versus common cancers) | 0.98 | 0.52-1.72 | 0.95 | NA | NA | NA |
| Number of lines of prior therapies ≥2 (versus <2) | 1.06 | 0.65-1.82 | 0.83 | NA | NA | NA |
| Number of metastatic sites ≥2 (versus <2) | 0.81 | 0.56-1.21 | 0.30 | NA | NA | NA |
| Liver metastases, yes (versus no) | 0.98 | 0.67-1.42 | 0.91 | NA | NA | NA |
| Brain metastases, yes (versus no) | 0.51 | 0.15-1.26 | 0.21 | NA | NA | NA |
| Pleural effusions or ascites, yes (versus no) | 0.31 | 0.13-0.64 | 0.004 | 0.37 | 0.15-0.78 | 0.015 |
| Suitable for biopsy, no (versus yes) | 0.53 | 0.32-0.83 | 0.009 | 0.50 | 0.30-0.80 | 0.006 |
| Neutrophils (/μl) <1500 (versus ≥1500) | 0.41 | 0.10-1.13 | 0.14 | 0.41 | 0.10-1.16 | 0.14 |
| Hemoglobin (g/dl) <9 (versus ≥9) | 0.25 | 0.06-0.67 | 0.019 | 0.49 | 0.12-1.42 | 0.25 |
| Platelets (×103/μl) <100 (versus ≥100) | 0.44 | 0.11-1.22 | 0.17 | 0.43 | 0.10-1.23 | 0.17 |
| Albumin (g/dl) <3.5 (versus ≥3.5) | 0.22 | 0.11-0.40 | <0.001 | 0.47 | 0.22-0.89 | 0.029 |
| Lactate dehydrogenase (U/l) ≥2.5 × ULN (versus <2.5 × ULN) | 0.15 | 0.02-0.48 | 0.008 | 0.26 | 0.04-0.90 | 0.043 |
| Creatinine (mg/dl) ≥1.5 (versus <1.5) | 0.52 | 0.03-2.60 | 0.53 | NA | NA | NA |
| Total bilirubin (mg/dl) ≥1.5 × ULN (versus <1.5 × ULN) | 0.98 | 0.05-5.43 | 0.99 | NA | NA | NA |
| Aspartate aminotransferase (U/l) ≥3.0 × ULN (versus <3.0 × ULN) | 0.25 | 0.01-1.17 | 0.17 | 0.30 | 0.02-1.55 | 0.25 |
| C-reactive protein (mg/dl) ≥5 (versus <5) | 0.20 | 0.05-0.54 | <0.001 | 0.44 | 0.10-1.30 | 0.19 |
| Place of residence, suburban/rural (versus urban) | 1.04 | 0.69-1.62 | 0.85 | NA | NA | NA |
| Referring hospital, community (versus hospital university hospital or cancer center) | 0.84 | 0.58-1.21 | 0.35 | NA | NA | NA |
| Median income (by zip code) <$25 000 (versus ≥$25 000) | 0.91 | 0.48-1.62 | 0.77 | NA | NA | NA |
| Genotype-nonmatched trial participation |
Univariable analysis |
Multivariable analysis |
||||
|---|---|---|---|---|---|---|
| Odds ratio | 95% Confidence interval | P | Odds ratio | 95% Confidence interval | P | |
| Travel costs ≥$100 (versus <$100) | 0.88 | 0.57-1.35 | 0.58 | 1.00 | 0.62-1.59 | 0.98 |
| Age ≥65 years (versus <65 years) | 1.02 | 0.69-1.52 | 0.89 | NA | NA | NA |
| Sex, women (versus men) | 0.79 | 0.53-1.16 | 0.23 | NA | NA | NA |
| Performance status ≥1 (versus 0) | 0.30 | 0.20-0.44 | <0.001 | 0.41 | 0.26-0.62 | <0.001 |
| Body mass index <18.5 (versus ≥18.5) | 0.60 | 0.32-1.03 | 0.083 | 0.86 | 0.44-1.56 | 0.63 |
| Tumor type, noncommon cancers (versus common cancers) | 0.49 | 0.20-1.03 | 0.078 | 0.44 | 0.17-0.96 | 0.047 |
| Number of lines of prior therapies ≥2 (versus <2) | 3.06 | 1.50-7.36 | 0.005 | 3.21 | 1.49-8.07 | 0.006 |
| Number of metastatic sites ≥2 (versus <2) | 0.67 | 0.45-1.02 | 0.052 | 0.64 | 0.41-1.02 | 0.059 |
| Liver metastases, yes (versus no) | 0.65 | 0.43-0.96 | 0.035 | 0.46 | 0.29-0.73 | <0.001 |
| Brain metastases, yes (versus no) | 0.58 | 0.17-1.43 | 0.30 | NA | NA | NA |
| Pleural effusions or ascites, yes (versus no) | 0.25 | 0.09-0.55 | 0.0026 | 0.33 | 0.11-0.78 | 0.021 |
| Suitable for biopsy, no (versus yes) | 0.41 | 0.23-0.68 | 0.001 | 0.27 | 0.14-0.48 | <0.001 |
| Neutrophils (/μl) <1500 (versus ≥1500) | 0.15 | 0.01-0.67 | 0.044 | 0.17 | 0.01-0.81 | 0.046 |
| Hemoglobin (g/dl) <9 (versus ≥9) | 0.09 | 0.004-0.39 | 0.015 | 0.24 | 0.01-1.17 | 0.16 |
| Platelets (×103/μl) <100 (versus ≥100) | 0.16 | 0.01-0.73 | 0.07 | 0.16 | 0.01-0.80 | 0.044 |
| Albumin (g/dl) <3.5 (versus ≥3.5) | 0.21 | 0.09-0.40 | <0.001 | 0.31 | 0.13-0.64 | 0.003 |
| Lactate dehydrogenase (U/l) ≥2.5 × ULN (versus <2.5 × ULN) | 0.32 | 0.08-0.87 | 0.046 | 0.46 | 0.11-1.34 | 0.21 |
| Creatinine (mg/dl) ≥1.5 (versus <1.5) | 0.97 | 0.61-1.50 | 0.98 | NA | NA | NA |
| Total bilirubin (mg/dl) ≥1.5 × ULN (versus <1.5 × ULN) | 0.92 | 0.62-1.40 | 0.97 | NA | NA | NA |
| Aspartate aminotransferase (U/l) ≥3.0 × ULN (versus <3.0 × ULN) | 0.84 | 0.46-1.62 | 0.97 | NA | NA | NA |
| C-reactive protein (mg/dl) ≥5 (versus <5) | 0.96 | 0.68-1.81 | 0.98 | NA | NA | NA |
| Place of residence, suburban/rural (versus urban) | 0.98 | 0.63-1.55 | 0.93 | NA | NA | NA |
| Referring hospital, community (versus hospital university hospital or cancer center) | 0.95 | 0.64-1.41 | 0.82 | NA | NA | NA |
| Median income (by zip code) <$25 000 (versus ≥$25 000) | 1.15 | 0.61-2.02 | 0.64 | NA | NA | NA |
All variables with P < 0.20 in the univariable analysis were entered into the multivariable analysis.
The significant values are shown in bold.
NA, not applicable; ULN, upper normal limit.
Figure 2.
Genotype-matched and genotype-nonmatched trial participation by travel costs. (A) Genotype-matched and genotype-nonmatched trial participation by travel costs. All variables with P < 0.20 in the univariable analysis (Table 1) were included in the multivariable analysis. (B, C) Travel cost map maps showing the NCCH at the center, marked with a blue circle. The green, red, and purple areas represent regions within a travel cost range of <$100, $100-200, and ≥$200 from the NCCH, respectively. The diagram was prepared using Google Maps and BioRender (https://www.biorender.com/). CI, confidence interval; NCCH, National Cancer Center Hospital; OR, odds ratio.
Association between travel costs and genotype-nonmatched trial participation
After adjusting for covariates, travel costs (≥$100 versus <$100) were not associated with the likelihood of genotype-nonmatched trial participation (9% versus 10%; OR = 0.62, 95% CI 0.62-1.59; P = 0.98; Table 2). Similarly, the likelihood of genotype-nonmatched participation did not decrease as travel costs increased from <$100 to $100-200 and ≥$200 (10% versus 9% versus 9%, respectively; OR = 1.01, 95% CI 0.49-1.96; OR = 1.00, 95% CI, 0.56-1.71; Figure 2A-C). In the sensitivity analysis, the OR for the complete-case analyses were consistent with those of the primary analysis.
Correlation between travel costs and travel distance/time
A high correlation was observed between travel costs and either travel time or distance among the overall patient population (Spearman’s correlation coefficient, 0.98, P < 0.0001; 0.98, P < 0.0001, respectively; Figure 3A and B). A similar high correlation was observed among subgroups of genotype-matched trial participants and genotype-nonmatched trial participants (genotype-matched trial participants: Spearman’s correlation coefficient; 0.98, P < 0.0001; 0.98, P < 0.0001, respectively; Figure 3C and D; genotype-nonmatched trial participants: Spearman’s correlation coefficient; 0.98, P < 0.0001; and 0.98, P < 0.0001, respectively; Figure 3E and F).
Figure 3.
Correlation between travel costs and travel distance/time. (A) Travel costs and distance among the overall patient population (n = 1127). (B) Travel costs and time among the overall patient population (n = 1127). (C) Travel costs and distance among genotype-matched trial participants (n = 114). (D) Travel costs and time among genotype-matched trial participants (n = 114). (E) Travel costs and distance among genotype-nonmatched trial participants (n = 127). (F) Travel costs and time among genotype-nonmatched trial participants (n = 127).
Relationship between travel burden and the time from the initial outpatient appointment to trial participation in the trial matching by genotype (genotype-matched versus genotype-nonmatched trial)
In genotype-matched trial participants, the time from the initial appointment to trial participation was significantly shorter than that in genotype-nonmatched trial participants (median: 21 versus 31 days, P = 0.006; Figure 4A). Disease progression was a more common reason for nonparticipation after screening in genotype-nonmatched trials than genotype-matched trials (P = 0.03; Figure 1 and Supplementary Table S3, available at https://doi.org/10.1016/j.esmorw.2025.100114). A low correlation was observed between travel burden (time, distance, or costs) and the time from the initial appointment to trial participation and overall trial participants (Spearman’s correlation coefficient: −0.026, P = 0.67; Spearman’s correlation coefficient: 0.009, P = 0.88; Spearman’s correlation coefficient: 0.011, P = 0.86, respectively; Figure 4B-D). A similar low correlation was observed among subgroups of genotype-matched trial and genotype-nonmatched trial participants (genotype-matched trial participants: Spearman’s correlation coefficient: −0.11, P = 0.19; Spearman’s correlation coefficient: −0.09, P = 0.33; Spearman’s correlation coefficient: −0.052, P = 0.56, respectively; Figure 4E-G; genotype-nonmatched trial participants: Spearman’s correlation coefficient: 0.071, P = 0.45; Spearman’s correlation coefficient: 0.11, P = 0.25; Spearman’s correlation coefficient: 0.067, P = 0.48, respectively; Figure 4H-J).
Figure 4.
Relationship between trial matching by genotype and travel burden, and the time from the initial appointment to trial participation. (A) Association between trial matching by genotype and the time from the initial outpatient appointment to trial participation among the overall trial participants (n = 241). P value was determined using the Mann–Whitney test. (B) Correlation between travel time and the time from the initial outpatient appointment to trial participation among the overall trial participants (n = 241). (C) Correlation between travel distance and the time from the initial appointment to trial participation among the overall trial participants (n = 241). (D) Correlation between travel costs and the time from the initial appointment to trial participation among the overall trial participants (n = 241). (E) Correlation between travel time and the time from the initial appointment to trial participation among genotype-matched trial participants (n = 114). (F) Correlation between travel distance and the time from the initial appointment to trial participation among genotype-matched trial participants (n = 114). (G) Correlation between travel costs and the time from the initial appointment to trial participation among genotype-matched trial participants (n = 114). (H) Correlation between travel time and the time from the initial appointment to trial participation among genotype-nonmatched trial participants (n = 127). (I) Correlation between travel distance and the time from the initial appointment to trial participation among genotype-nonmatched trial participants (n = 127). (J) Correlation between travel costs and the time from the initial appointment to trial participation among genotype-nonmatched trial participants (n = 127). IQR, interquartile range.
Discussion
Our results showed that increased travel costs were significantly associated with a decreased likelihood of genotype-matched trial participation. However, this association was not observed in genotype-nonmatched trial participants. Travel burden (time, distance, or costs) showed a low correlation with the time from the initial appointment to trial participation in both genotype-matched and genotype-nonmatched trials. In genotype-matched trial participants, the time from the initial appointment to trial participation was significantly shorter than that in genotype-nonmatched trial participants.
To the best of our knowledge, this is the first research to investigate the association between travel costs and participation in genotype-matched trials following CGP testing. In our research, patients with travel costs ≥$100 (round trip) were significantly less likely to participate in genotype-matched trials than those with travel costs <$100. While payments for travel costs to patients for their trial participation are common, the payment is usually fixed for each trial site in Japan, regardless of the travel distance. Moreover, the $100 threshold in our research represented 5% of the average monthly income in Japan, posing a significant financial burden.35 In 2018, the Food and Drug Administration (FDA) allowed reimbursement of travel costs to encourage participation from rural areas.36 Implementing adequate financial support and decentralizing clinical trials (DCTs) could alleviate these financial travel burdens. In the United States, supported by the FDA, DCTs can operate without a fixed site and managed remotely by telemedicine or provided by the patient’s local oncologist.37, 38, 39 Our hospital is conducting the first DCT in Japan, utilizing a satellite institution for a company-sponsored phase I clinical trial in oncology.40
We found low correlations between travel burden (time, distance, and cost) and time from the initial clinic appointment to trial participation in both genotype-matched and genotype-nonmatched trials. These findings indicate that the travel burden does not significantly delay the enrollment process. Conversely, the time from initial appointment to participation in genotype-matched trials was significantly shorter than that in genotype-nonmatched trials; hence genotype-matched trials fill up more quickly, underscoring the importance of rapid shared decision-making.37 Trial enrollment tends to be encouraged in genotype-matched trials because genotype-matched therapies are more likely to have antitumor effects than genotype-nonmatched therapies.41, 42, 43, 44, 45, 46
Our findings revealed that, in contrast to genotype-matched trials, travel costs did not affect the likelihood of participation in genotype-nonmatched trials. The longer time frames associated with genotype-nonmatched trial participation could lead to disease progression, including uncontrolled liver metastases, pleural effusions, or ascites, significantly influencing the probability of participation. In fact, in our research, disease progression was a more commonly cited reason for screening failure in genotype-nonmatched than genotype-matched trials. In addition, genotype-matched trials in our study predominantly consisted of phase I trials (65%) and targeted drug monotherapy (69%). These trials typically have faster recruitment timelines and require more frequent visits for safety monitoring after enrollment, often during morning hours. Moreover, the relationship between travel costs and individual income might be a significant factor.9,47 While we gathered household income data, our research did not specifically examine actual individual income.
This research has some limitations. First, it was conducted at a single cancer center in Japan, which limits its generalizability. However, the proportion of patients receiving genotype-matched therapies after undergoing CGP testing is comparable to that in the United States (10%-20%).14, 15, 16, 17, 18 Second, the estimation of travel costs did not include transportation by personal vehicle, despite the public transportation system being readily accessible to most Japanese patients. Finally, socioeconomic factors, such as family or marital status, occupation, and educational background, were not evaluated, which could potentially confound trial participation.
In conclusion, increased travel costs were associated with a decreased likelihood of participation in genotype-matched trials following CGP testing and referral to the NCCH. In genotype-matched trial participants, the time from the initial appointment to trial participation was significantly shorter than that in genotype-nonmatched trial participants. Our findings suggest that financial burdens due to travel cause inequities in precision oncology, which can be mitigated by DCT or financial assistance programs.
Acknowledgements
The authors thank R. Tsunoda for assistance with data collection.
Funding
None declared.
Disclosure
TK has received personal fees from Chugai, AstraZeneca, and Sysmex, as well as grants from PACT Pharma, Chugai, Daiichi-Sankyo, Novartis, Eli Lilly, Pfizer, Janssen Pharma, Zymeworks, Noile, and Takeda outside the submitted work. JS has received personal fees from Eisai, Amgen, and Chugai Pharmaceutical Co. outside the submitted work. KS has received personal fees from AstraZeneca and Pfizer outside the submitted work. SK has received research funding from ASLAN Pharmaceuticals, AstraZeneca, Bayer, Eli Lilly, MSD, BI, and Pfizer outside the submitted work. TY has received grants and personal fees from Amgen, AstraZeneca, Ono, Merck Sharp & Dohme, Novartis, Chugai, and Bristol-Myers Squibb; grants from Takeda, Daiichi Sankyo, and AbbVie; and personal fees from Taiho, Eli Lilly, Roche, and ArcherDX outside of the submitted work. HS has received grants from Ono pharmaceutical Co., Ltd., MSD, Astellas, and Amgen outside the submitted work. KY has received personal fees from Eisai, Pfizer, AstraZeneca, Takeda, Taiho, and Chugai outside the submitted work. NY has received grants from Chugai, Taiho, Eisai, Eli Lilly, Quintiles, Astellas, Bristol-Myers Squibb, Novartis, Daiichi Sankyo, Pfizer, Boehringer Ingelheim, Kyowa-Hakko Kirin, Bayer, Ono Pharmaceutical Co., Ltd., Takeda, Janssen Pharma, Merck Sharp & Dohme, Merck, GlaxoSmithKline, Sumitomo Dainippon, Chiome Bioscience Inc., Otsuka, Carna Biosciences, Genmab, and Shionogi; and personal fees from Ono Pharmaceutical Co., Ltd., Chugai, AstraZeneca, Pfizer, Lilly, Bristol-Myers Squibb, Eisai, Otsuka, Takeda, Boehringer Ingelheim, Cimic, Sysmex, and Eisai, outside of the submitted work. All other authors have declared no conflicts of interest.
Data sharing
The data that support the findings of this study are available from the corresponding author on reasonable request.
Supplementary data
References
- 1.Chihara D., Lin R., Flowers C.R., et al. Early drug development in solid tumours: analysis of National Cancer Institute-sponsored phase 1 trials. Lancet. 2022;400(10351):512–521. doi: 10.1016/S0140-6736(22)01390-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bittlinger M., Bicer S., Peppercorn J., Kimmelman J. Ethical considerations for phase I trials in oncology. J Clin Oncol. 2022;40(30):3474–3488. doi: 10.1200/JCO.21.02125. [DOI] [PubMed] [Google Scholar]
- 3.Wercholuk A.N., Parikh A.A., Snyder R.A. The road less traveled: transportation barriers to cancer care delivery in the rural patient population. JCO Oncol Pract. 2022;18(9):652–662. doi: 10.1200/OP.22.00122. [DOI] [PubMed] [Google Scholar]
- 4.Pancoe S., Ko E.M., Smith A.J.B. Federal transportation regulations limit access to clinical trials and oncology care. JAMA Oncol. 2024;10(4):425–426. doi: 10.1001/jamaoncol.2023.6767. [DOI] [PubMed] [Google Scholar]
- 5.Uehara Y., Koyama T., Katsuya Y., et al. Travel time and distance and participation in precision oncology trials at the National Cancer Center Hospital. JAMA Netw Open. 2023;6(9) doi: 10.1001/jamanetworkopen.2023.33188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rocque G.B., Williams C.P., Miller H.D., et al. Impact of travel time on health care costs and resource use by phase of care for older patients with cancer. J Clin Oncol. 2019;37(22):1935–1945. doi: 10.1200/JCO.19.00175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Borno H.T., Zhang L., Zhang S., et al. Implementation of a multisite financial reimbursement program in cancer clinical trials integrated with patient navigation: a pilot randomized clinical trial. JCO Oncol Pract. 2022;18(6):e915–e924. doi: 10.1200/OP.21.00328. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Meropol N.J., Buzaglo J.S., Millard J., et al. Barriers to clinical trial participation as perceived by oncologists and patients. J Natl Compr Cancer Netw. 2007;5(8):655–664. doi: 10.6004/jnccn.2007.0067. [DOI] [PubMed] [Google Scholar]
- 9.Unger J.M., Gralow J.R., Albain K.S., et al. Patient income level and cancer clinical trial participation: a prospective survey study. JAMA Oncol. 2016;2(1):137–139. doi: 10.1001/jamaoncol.2015.3924. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Borno H.T., Zhang L., Siegel A., Chang E., Ryan C.J. At what cost to clinical trial enrollment? A retrospective study of patient travel burden in cancer clinical trials. Oncologist. 2018;23(10):1242–1249. doi: 10.1634/theoncologist.2017-0628. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Nipp R.D., Powell E., Chabner B., Moy B. Recognizing the financial burden of cancer patients in clinical trials. Oncologist. 2015;20(6):572–575. doi: 10.1634/theoncologist.2015-0068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Paydary K., DeLuca A., Aggarwal R., Wall L., Stadler W.M. Enrollment barriers for molecular targeted trials. JAMA Oncol. 2023;9(6):863–864. doi: 10.1001/jamaoncol.2023.0248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Unger J.M., Shulman L.N., Facktor M.A., Nelson H., Fleury M.E. National estimates of the participation of patients with cancer in clinical research studies based on commission on cancer accreditation data. J Clin Oncol. 2024;42(18):2139–2148. doi: 10.1200/JCO.23.01030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Meric-Bernstam F., Brusco L., Shaw K., et al. Feasibility of large-scale genomic testing to facilitate enrollment onto genomically matched clinical trials. J Clin Oncol. 2015;33(25):2753–2762. doi: 10.1200/JCO.2014.60.4165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Zehir A., Benayed R., Shah R.H., et al. Mutational landscape of metastatic cancer revealed from prospective clinical sequencing of 10,000 patients. Nat Med. 2017;23(6):703–713. doi: 10.1038/nm.4333. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Ida H., Koyama T., Mizuno T., et al. Clinical utility of comprehensive genomic profiling tests for advanced or metastatic solid tumor in clinical practice. Cancer Sci. 2022;113(12):4300–4310. doi: 10.1111/cas.15586. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Stockley T.L., Oza A.M., Berman H.K., et al. Molecular profiling of advanced solid tumors and patient outcomes with genotype-matched clinical trials: the princess Margaret IMPACT/COMPACT trial. Genome Med. 2016;8(1):109. doi: 10.1186/s13073-016-0364-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Johnson A., Khotskaya Y.B., Brusco L., et al. Clinical use of precision oncology decision support. JCO Precis Oncol. 2017;2017 doi: 10.1200/PO.17.00036. PO.17.00036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Sunami K., Ichikawa H., Kubo T., et al. Feasibility and utility of a panel testing for 114 cancer-associated genes in a clinical setting: a hospital-based study. Cancer Sci. 2019;110(4):1480–1490. doi: 10.1111/cas.13969. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Frampton G.M., Fichtenholtz A., Otto G.A., et al. Development and validation of a clinical cancer genomic profiling test based on massively parallel DNA sequencing. Nat Biotechnol. 2013;31(11):1023–1031. doi: 10.1038/nbt.2696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Woodhouse R., Li M., Hughes J., et al. Clinical and analytical validation of FoundationOne liquid CDx, a novel 324-gene cfDNA-based comprehensive genomic profiling assay for cancers of solid tumor origin. PLoS One. 2020;15(9) doi: 10.1371/journal.pone.0237802. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kim S.T., Banks K.C., Lee S.H., et al. Prospective feasibility study for using cell-free circulating tumor DNA-guided therapy in refractory metastatic solid cancers: an interim analysis. JCO Precis Oncol. 2017;1 doi: 10.1200/PO.16.00059. PO.16.0005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Naito Y., Sunami K., Kage H., et al. Concordance between recommendations from multidisciplinary molecular tumor boards and central consensus for cancer treatment in Japan. JAMA Netw Open. 2022;5(12) doi: 10.1001/jamanetworkopen.2022.45081. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kohno T., Kato M., Kohsaka S., et al. C-CAT: the national datacenter for cancer genomic medicine in Japan. Cancer Discov. 2022;12(11):2509–2515. doi: 10.1158/2159-8290.CD-22-0417. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Tamura T., Ikegami M., Kanemasa Y., et al. Selection bias due to delayed comprehensive genomic profiling in Japan. Cancer Sci. 2023;114(3):1015–1025. doi: 10.1111/cas.15651. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Sunami K., Naito Y., Saigusa Y., et al. A learning program for treatment recommendations by molecular tumor boards and artificial intelligence. JAMA Oncol. 2024;10(1):95–102. doi: 10.1001/jamaoncol.2023.5120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Gerber D.E., Singh H., Larkins E., et al. A new approach to simplifying and harmonizing cancer clinical trials-standardizing eligibility criteria. JAMA Oncol. 2022;8(9):1333–1339. doi: 10.1001/jamaoncol.2022.1664. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Tan T., Rheaume M., Wang L., et al. Referrals to a phase I clinic and trial enrollment in the molecular screening era. Oncologist. 2019;24(7):e518–e525. doi: 10.1634/theoncologist.2018-0808. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Sung H., Ferlay J., Siegel R.L., et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209–249. doi: 10.3322/caac.21660. [DOI] [PubMed] [Google Scholar]
- 30.Statistics Bureau of Japan Population Census. 2020. https://www.stat.go.jp/english/data/kokusei/index.html Available at.
- 31.Arkenau H.T., Barriuso J., Olmos D., et al. Prospective validation of a prognostic score to improve patient selection for oncology phase I trials. J Clin Oncol. 2009;27(16):2692–2696. doi: 10.1200/JCO.2008.19.5081. [DOI] [PubMed] [Google Scholar]
- 32.Wheler J., Tsimberidou A.M., Hong D., et al. Survival of 1,181 patients in a phase I clinic: the MD Anderson Clinical Center for targeted therapy experience. Clin Cancer Res. 2012;18(10):2922–2929. doi: 10.1158/1078-0432.CCR-11-2217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.van Buuren S., Groothuis-Oudshoorn K. mice: Multivariate imputation by chained equations in R. J Stat Softw. 2011;45(3):1–67. [Google Scholar]
- 34.Mukaka M.M. Statistics corner: a guide to appropriate use of correlation coefficient in medical research. Malawi Med J. 2012;24(3):69–71. [PMC free article] [PubMed] [Google Scholar]
- 35.National Tax Agency JAPAN Tax statistics. 2021. https://www.nta.go.jp/publication/statistics/kokuzeicho/tokei.htm Available at.
- 36.Food and Drug Administration (FDA) Payment and reimbursement to research subjects: guidance for institutional review boards and clinical investigators. January 1, 2018. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/payment-and-reimbursement-research-subjects Available at.
- 37.Nipp R.D., Hong K., Paskett E.D. Vol. 39. Am Soc Clin Oncol Educ Book; 2019. pp. 105–114. (Overcoming barriers to clinical trial enrollment). [DOI] [PubMed] [Google Scholar]
- 38.Harvey R.D., Miller T.M., Hurley P.A., et al. A call to action to advance patient-focused and decentralized clinical trials. Cancer. 2024;130(8):1193–1203. doi: 10.1002/cncr.35145. [DOI] [PubMed] [Google Scholar]
- 39.Underhill C., Freeman J., Dixon J., et al. Decentralized clinical trials as a new paradigm of trial delivery to improve equity of access. JAMA Oncol. 2024;10(4):526–530. doi: 10.1001/jamaoncol.2023.6565. [DOI] [PubMed] [Google Scholar]
- 40.National Cancer Center Japan Chugai, NCCH, OMPU and MICIN start company-sponsored phase I study in oncology with a new decentralized clinical trial structure. National Cancer Center Japan. http://www.ncc.go.jp/en/information/press_release/20200508/index.html Available at.
- 41.Sicklick J.K., Kato S., Okamura R., et al. Molecular profiling of cancer patients enables personalized combination therapy: the I-PREDICT study. Nat Med. 2019;25(5):744–750. doi: 10.1038/s41591-019-0407-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Sicklick J.K., Kato S., Okamura R., et al. Molecular profiling of advanced malignancies guides first-line N-of-1 treatments in the I-PREDICT treatment-naïve study. Genome Med. 2021;13(1):155. doi: 10.1186/s13073-021-00969-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Adashek J.J., Subbiah V., Westphalen C.B., Naing A., Kato S., Kurzrock R. Cancer: slaying the nine-headed Hydra. Ann Oncol. 2023;34(1):61–69. doi: 10.1016/j.annonc.2022.07.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Uehara Y., Ikeda S., Kim K.H., et al. Targeting the FGF/FGFR axis and its co-alteration allies. ESMO Open. 2022;7(6) doi: 10.1016/j.esmoop.2022.100647. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Gouda M.A., Buschhorn L., Schneeweiss A., Wahida A., Subbiah V. N-of-1 trials in cancer drug development. Cancer Discov. 2023;13(6):1301–1309. doi: 10.1158/2159-8290.CD-22-1377. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Kato S., Kim K.H., Lim H.J., et al. Real-world data from a molecular tumor board demonstrates improved outcomes with a precision N-of-One strategy. Nat Commun. 2020;11(1):4965. doi: 10.1038/s41467-020-18613-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Lee H., Bates A.S., Callier S., et al. Analysis and optimization of equitable US cancer clinical trial center access by travel time. JAMA Oncol. 2024;10(5):652–657. doi: 10.1001/jamaoncol.2023.7314. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.