Impact of Clinical Practice Gaps on the

special articles Implementation of Personalized Medicine in

Advanced Non–Small-Cell Lung Cancer
Helen Sadik, PhD1; Daryl Pritchard, PhD2; Derry-Mae Keeling, BSc1; Frank Policht, PhD1; Peter Riccelli, PhD1; Gretta Stone, BS3;
Kira Finkel, MSPH3; Jeff Schreier, MBA1; and Susanne Munksted, MS1
abstract

PURPOSE Personalized medicine presents new opportunities for patients with cancer. However, many patients
do not receive the most effective personalized treatments because of challenges associated with integrating
predictive biomarker testing into clinical care. Patients are lost at various steps along the precision oncology
pathway because of operational inefficiencies, limited understanding of biomarker strategies, inappropriate
testing result usage, and access barriers. We examine the impact of various clinical practice gaps associated
with diagnostic testing-informed personalized medicine strategies on the treatment of advanced non–small-cell
lung cancer (aNSCLC).
METHODS Using Diaceutics’ Data Repository, a multisource database including commercial and Medicare
claims and laboratory data from over 500,000 patients with non–small-cell lung cancer in the United States, we
analyzed the number of patients with newly diagnosed aNSCLC who could have, but did not, benefit from a
personalized treatment. The analysis focuses on the independent and cumulative impacts of gaps occurring
during seven steps of the precision oncology pathway, from diagnosis to treatment.
RESULTS For every 1,000 patients in the study cohort, 497 (49.7%) are lost to precision oncology because of
factors associated with getting biomarker test results. Among the 503 of 1,000 patients who did receive results
from a biomarker test, 147 (29.2%) did not receive appropriate targeted treatments. Thus, approximately 64% of
potentially eligible patients with aNSCLC are not benefiting from precision oncology therapies appropriate for
their disease.
CONCLUSION Most patients with aNSCLC eligible for precision oncology treatments do not benefit from them
because of clinical practice gaps. This finding is likely reflective of similar gaps in other cancer types. An
increased understanding of the impact of each practice gap can inform strategies to improve the delivery of
precision oncology, helping to fully realize the promise of personalized medicine.
JCO Precis Oncol 6:e2200246. © 2022 by American Society of Clinical Oncology
Creative Commons Attribution Non-Commercial No Derivatives 4.0 License

INTRODUCTION subsequently used to guide therapeutic decisions. Bio-

Precision oncology strategies are an important and marker testing and targeted therapeutics are relatively
growing component of cancer care. There are over 90 US new, however, and providers face several challenges in
ASSOCIATED
CONTENT Food and Drug Administration–approved targeted ther- adapting cancer care practices accordingly. Levels of
biomarker testing differ greatly across practice settings,
Data Supplement apies available for use in eligible patients with cancer,1
Author affiliations
tumor types, and biomarkers, with varying adherence to
and a recent oncology pipeline report showed that ap-
and support testing guidelines.3 Implementation challenges are ex-
proximately 55% of all oncology clinical trials involved the
information (if emplified in non–small-cell lung cancer (NSCLC), where
applicable) appear at use of biomarkers.2 Predictive biomarker testing to help although more than 70% of patients have tumors with
the end of this identify patients who could benefit from targeted thera- biomarker alterations related to therapeutic options,4
article. pies is a cornerstone of personalized medicine in cancer many patients still do not receive biomarker testing.3,5
Accepted on August care, allowing for more rapid diagnosis while informing
31, 2022 and Furthermore, many cancer patients with actionable
published at treatment decisions that could lead to better patient
results as determined by biomarker testing do not
ascopubs.org/journal/ outcomes and systemic efficiencies.
actually receive appropriate targeted therapies. Re-
po on October 31,
2022: DOI https://doi.
The success of precision oncology relies on the ac- ports show that more than one third of US patients with
org/10.1200/PO.22. curate identification of patients harboring biomarker cancer miss out on precision oncology treatment
00246 alterations as determined by laboratory test results because of suboptimal testing practices specifically

1
Downloaded from ascopubs.org by 125.165.106.29 on February 23, 2023 from 125.165.106.029
Copyright © 2023 American Society of Clinical Oncology. All rights reserved.
 Sadik et al

CONTEXT
Key Objective
This article reflects a large multisource US commercial and Medicare claims and laboratory database analysis of the number of
patients with newly diagnosed advanced non–small-cell lung cancer (NSCLC) who could have, but did not, benefit from a
personalized treatment because of various clinical practice gaps occurring during the delivery of precision oncology care,
from diagnosis to treatment.
Knowledge Generated
We examine clinical practice gaps along the precision NSCLC treatment pathway and quantify the loss of patients because of
each clinical gap, including preanalytic biomarker testing and post-analytic practice challenges, ultimately showing that
approximately 64% of potentially eligible patients with advanced NSCLC are not benefiting from precision oncology
therapies appropriate for their disease.
Relevance
Improving the implementation of precision oncology care requires a better understanding of the associated clinical practice
gaps and their impact on patient care.

related to quality or sample management issues.6,7 For clinical practice gaps associated with diagnostic testing-
NSCLC, studies examining practice-based data from over informed personalized medicine strategies in NSCLC and
190 US hospital systems have estimated that only 65%- quantified the loss of patients along the precision oncology
75% of patients with an actionable mutation actually pathway because of each practice gap.
receive targeted therapies.8-10 Another study examining
patients with NSCLC within the Veterans Affairs National METHODS
Precision Oncology Program revealed that more than 30% To quantify the extent to which clinical practice gaps are
of patients with highly actionable gene variants received affecting the treatment of patients with advanced NSCLC
chemotherapy instead of more effective targeted (aNSCLC) in the US health care setting, data from a large
treatments.11 population of newly diagnosed patients were analyzed to
estimate the number of patients who could have, but did
Testing and treatment misfires in precision oncology have
not, benefit from a personalized treatment. The analysis
clinical consequences. An examination of a large registry
focused on the independent and cumulative impact of the
claims database from 2010 through 2018, for example,
gaps occurring during seven discrete steps of the precision
showed that a significant percentage of patients with ad-
oncology pathway from diagnosis to treatment. These
vanced epidermal growth factor receptor–positive (EGFR+)
practice gaps encompass barriers observed at each step,
and anaplastic lymphoma kinase–positive (ALK+) NSCLC
broadly summarized as follows:
who did not receive targeted therapies had inferior survival
rates.12 • Step 1: Biopsy referral: Initial solid or blood biopsy was
never performed.
These inefficiencies may exemplify the clinical challenges
• Step 2: Biospecimen collection: Biospecimen collec-
of navigating the complex precision oncology pathway,
tion challenges including insufficient tissue or tumor
which includes multiple steps from diagnosis to treatment.
cell content of initial biopsy or rebiopsy inhibited
At various steps, clinical practice gaps caused by opera-
biomarker testing and its accuracy.
tional inefficiencies, limited awareness or understanding of
• Step 3: Biospecimen evaluation/pathology: Biospeci-
biomarker strategies, inappropriate use of testing results,
men tumor cell content was overestimated, inhibiting
and coverage and payment challenges can lead to missed
biomarker testing and its accuracy.
opportunities for patients to benefit from targeted treat-
• Step 4: Biomarker test ordering: Appropriate testing
ments. These clinical practice gaps can occur during
was not ordered, or treatment began before testing
preanalytic stages of diagnostic testing related to biopsy
was ordered.
collection and evaluation and test ordering. They can also
• Step 5: Biomarker testing performance: Biomarker
occur during analytical and postanalytic stages related to
testing provided inconclusive or false-negative (FN)
test performance, result reporting, and treatment decisions.
results.
Implementing consistent biomarker testing-informed pre- • Step 6: Test result reporting: As a result of turnaround
cision oncology care requires a better understanding of the time (TAT) delays, treatment was initiated without
associated clinical practice gaps and the impact each gap consideration of test results.
has on patient care. In this study, we used laboratory and • Step 7: Treatment decision: Targeted treatment was
claims-based data from the US health system to examine not selected despite positive test results.

2 © 2022 by American Society of Clinical Oncology

Downloaded from ascopubs.org by 125.165.106.29 on February 23, 2023 from 125.165.106.029

Copyright © 2023 American Society of Clinical Oncology. All rights reserved.
 Clinical Practice Gaps Affecting Precision Oncology

Data Source and Population was not determined, the data sets were correlated to make
Analysis was based on Diaceutics’ proprietary DXRX Data population-level assessments.
Repository, a multisource database that consists of com- Data Supplement includes detailed information, broken
mercial and Medicare claims and laboratory data. The data down by practice gap, about how data from the Diaceutics
set contains real-time laboratory data, deidentified at a pa- Data Repository were combined with evidence from other
tient level, and covers 340 million lives. The data repository sources to estimate the relative impact each gap contrib-
was developed in 2015 and is continuously updated weekly utes to the number of patients who could have but did not
and is funded by Diaceutics.13 Data contain patient diagnosis benefit from personalized medicine.
code; physician national provider identifier; laboratory- Clinical Practice Gap Analysis
performed, test and panel name; result of test (positive,
negative, quantity not sufficient, test not performed, in- The extent to which clinical practice gaps at each of the seven
conclusive, etc); and any specific genomic alterations (eg, steps along the precision oncology pathway contributes to the
mutations, fusions, etc) detected. The repository covers 84% overall number of patients who could have, but did not, benefit
of patients with lung cancer whose data are included in the from receiving personalized medicine was quantified through
US National Cancer Institute’s SEER Program database.14 an integrated approach using Diaceutics’ Data Repository
along with additional sources of publicly available data as
Within the Medicare claims portion of the repository, a described above. The step-by-step approach and the number
population of 38,068 patients with aNSCLC newly diag- of patients who were potentially available for analysis at each
nosed and actively managed in 2019 were identified for step of the journey are shown in Figure 1.
practice gap analyses. Demographics of the patients in-
cluded in this study are summarized in Table 1. Actively RESULTS
managed patients were defined as having three or more Data from 506,889 patients with NSCLC were included within
appearances/procedures noted in claims data, thereby the Diaceutics Data Repository in 2019. Overall, 144,486
providing enough data to construct a clear picture of their cases were first-time NSCLC insurance claims, indicating
clinical journeys. Patients were further categorized on the patients with newly diagnosed NSCLC. Of them, 38,068 were
basis of whether they had a biopsy performed (tissue or determined to be patients with actively managed aNSCLC. Of
liquid); their biomarker testing methodology (immunohis- them, we evaluated the number and percentage of patients
tochemistry [IHC], fluorescence in situ hybridization, next- who advanced or were lost at each step within the clinical
generation sequencing [NGS], Sanger sequencing, or practice gap framework. We normalized the data to a patient
polymerase chain reaction [PCR]); and their treatment population of 1,000 to easily demonstrate the percentage of
category (chemotherapy, immunotherapy, or targeted). eligible patients who may be lost to receiving targeted ther-
Patients who were diagnosed or underwent testing in apies because of each clinical practice gap (Fig 2).
the final 3 months of 2019 often had incomplete treatment Clinical Practice Gap 1: Initial Biopsy Was
data and were removed from treatment analysis. The Never Performed
type of therapy received was evaluated using claims as-
sociated with Medicare Parts A/B (inpatient/outpatient/ Of the 38,068 patients with actively managed aNSCLC, our
non–hospital-based care provision) and D (prescription drugs). analysis reveals that 6.6% never had an initial biopsy
performed (84.6% received a tissue biopsy; 8.8% received
In the analyses of practice gaps 1 and 7, additional a liquid biopsy, 6.6% of patients without tissue or liquid
published data related to patients with aNSCLC statistics biopsy were likely diagnosed through imaging only).
were used to confirm or supplement Diaceutics’ Therefore, clinical practice gap 1 has led to 66 of 1,000
repository.14-16 In the analysis of practice gap 2, additional patients lost. The remaining 934 patients advance on the
published data related to tissue and tumor cell sufficiency precision oncology care pathway.
rates were used.17-19 In the analyses of practice gaps 3 and
6, additional published data were used related to meth- Clinical Practice Gap 2: Biospecimen Collection
odology dependent analysis rates.20-25 In the analyses of Challenges Inhibited Biomarker Testing
practice gaps 5 and 7, additional published data related to Of the 32,224 patients with aNSCLC who received a tissue
methodology dependent FN, false-positive, or true-positive biopsy, data on the biopsy type were available for 9,425
rates were used.15,20,22,25-35 Furthermore, in the analyses of cases, with 79.8% being conducted using fine needle as-
practice gaps 5 and 7, additional third-party real-time pirates (FNAs), 9.9% using core-needle biopsies (CNBs),
laboratory data from US-based laboratories were used to and 10.3% involving surgical resections. We applied an
supplement the Medicare claims data and laboratory test insufficient tissue collected from biopsy rate of 4.5%17 to the
results data (eg, test performance and positivity rates) in- 89.7% of tissue biopsy samples including FNA plus CNB for
cluded in the Diaceutics repository. A total of 5,589 patients an estimated 4.0% rate of patients with tissue insufficiency of
with aNSCLC newly diagnosed in 2019 were identified initial biopsy to continue with biomarker testing. Liquid bi-
within the real-time laboratory data set. Although the direct opsies and surgical resections were assumed to have suf-
overlap between the patients included in the two data sets ficient genetic material for biomarker testing.

JCO Precision Oncology 3

Downloaded from ascopubs.org by 125.165.106.29 on February 23, 2023 from 125.165.106.029

Copyright © 2023 American Society of Clinical Oncology. All rights reserved.
 Sadik et al

TABLE 1. Demographics of the Patients With Newly Diagnosed TABLE 1. Demographics of the Patients With Newly Diagnosed
Advanced Non–Small-Cell Lung Cancer Included in This Analysis Advanced Non–Small-Cell Lung Cancer Included in This Analysis
(on the basis of claims data in 2019) (on the basis of claims data in 2019) (Continued)
Medicare Patient Demographics Medicare Patient Demographics

Patient Patient
Volume Breakdown Volume Breakdown
n = 38,068 (%) n = 38,068 (%)
Sex Primary physician specialty
Male 49.20 Oncologist 69.90
Female 50.80 Surgery 14.14
Age, years Internal medicine 7.09
30 or younger 0.00 Pathologist 5.55
31-40 0.20 Radiologist 1.12
41-50 0.80 Specialist 0.71
a
51-60 4.70 Others 1.50
61-70 29.70
Abbreviation: COPD, chronic obstructive pulmonary disease.
71-80 40.00 a
Others category includes student in health care training/education,
81-90 20.70 nurse practitioner, physician assistant, legal medicine, family
medicine, pediatrics, hospitalist, neurology, urology, emergency
91 or older 3.90
medicine, otolaryngology, general acute care hospital, obstetrics &
Race gynecology, dermatology, nuclear medicine, and general practice.
Non-Hispanic White 84.50
Of the 9,425 patients who received a tissue biopsy of known
Non-Hispanic Black 9.30
type, 5.9% received a rebiopsy for various reasons. We
Others 1.90 applied an insufficient tissue rate for rebiopsy of 13.6%18 to
Non-Hispanic Asian/Pacific Islander 1.40 2.7% of the overall samples (% of patients with FNA plus
Hispanic 1.30 CNB rebiopsy) for an estimated 0.37% rate of patients who
Non-Hispanic American Indian/Alaska 1.20 were unable to continue testing. Additionally, 0.6% of
Native patients were lost who had original tissue biopsy collection
Unknown 0.50 problems but were not able to be rebiopsied for various
US regions
reasons such as advanced disease progression or tumor
inaccessibility and as a result did not go on to receive
South 40.70
biomarker testing, resulting in a loss of a total of 0.97% of
Midwest 24.00 patients with tissue insufficiency of rebiopsy.
Northeast 19.40
Of the patients with sufficient amount of tissue from biopsy
West 15.80
or rebiopsy, an insufficient number of tumor cells available
Comorbidity for testing within the tissue sample was estimated to be a
Hypertension 71.70 problem for 10.7% of patients.19 We applied the 10.7% rate
Hyperlipidemia 57.40 of tumor cell unavailability to the 89.7% of samples ob-
COPD 51.30
tained via FNA or CNB to determine a 9.6% rate of patients
unable to receive testing because of the absence of suf-
Anemia 48.00
ficient tumor cells.
Ischemic heart disease 46.70
Thus, overall, practice gap 2 has led to 136 of 934 patients
Chronic kidney disease 39.70
lost. The remaining 798 patients advance on the precision
Diabetes 31.00
oncology care pathway.
Heart failure 27.40
Clinical Practice Gap 3: Biopsy Specimen Tumor Cell
Hypothyroidism 19.50
Content Was Overestimated, Inhibiting Biomarker Testing
Prostatic hyperplasia 12.90
and Its Accuracy
Stroke/tia 7.50
Of the patients with aNSCLC who had adequate tumor
Asthma hyperlipidemia 7.10
biopsy samples, we estimate that 1.7% had tumor cell
Acute myocardial infarction 2.30 content that was overestimated. This was calculated as
(Continued in next column) follows: 14% of tissues had , 20% tumor content and,

4 © 2022 by American Society of Clinical Oncology

Downloaded from ascopubs.org by 125.165.106.29 on February 23, 2023 from 125.165.106.029

Copyright © 2023 American Society of Clinical Oncology. All rights reserved.
 Clinical Practice Gaps Affecting Precision Oncology

Biomarker test
ordering (test not
ordered; gap 4)
Biomarker test ordering
Biospecimen (test ordered after
evaluation/pathology treatment; gap 4) Treatment decision
(gap 3) (inappropriate, unmatched
or no treatment; gap 7)
Biospecimen Test reporting
collection – (turnaround time; gap 6)
rebiopsy (gap 2)

Biospecimen Biomarker testing

Biopsy referral performance Patients tested for any of
collection –
(gap 1) (gap 5) the following activating
initial biopsy (gap 2) Patients
mutations (ALK, BRAF, received
EGFR, KRAS, NTRK, and
appropriate
Patients with a ROS1), potentially treatment
biomarker test Patients with eligible for TKI
Patients with newly biomarker test and (n = 24,187)
Patients with a biopsy (inclusive of IHC,
diagnosed aNSCLC treatment decision
(FNA, CNB, surgical FISH, NGS, Sanger
in Medicare claims, within study
Claims resection or liquid sequencing, RT-PCR,
actively managed timeframe limitations
data biopsy; n = 35,556) and ddPCR
(n = 38,068) (n = 27,186)
methodologies;
n = 29,227) Patients
Patients tested for
received
Diaceutics Data PD-L1, potentially eligible
appropriate
Lab Repository Prescription for IO (n = 23,627)
Data used to complement treatment
demographics data claim analysis with result-
(N = 506,889; 2019)
level details

Patients tested for

one of the following
Laboratory Patients with aNSCLC
with result reported:
data in real-time Treatment decision
ALK, BRAF, EGFR,
laboratory data (inappropriate, unmatched
KRAS, MET, NTRK,
(n = 5,589; 2019) or no treatment; gap 7)
PD-L1, RET, or ROS1
(n = 4,885)

Biomarker testing Treatment decision

performance (false-positive rate; gap 7)
(gap 5)

Treatment
decision
(positivity rate; gap 7)

FIG 1. Step-by-step approach for analyzing the number of patients with advanced non–small-cell lung cancer lost at each step of the precision
oncology pathway. CNB, core-needle biopsies; ddPCR, droplet digital polymerase chain reaction; FISH, fluorescence in situ hybridization; FNA, fine-
needle aspirates; IHC, immunohistochemistry; IO, immuno-oncology; aNSCLS, advanced non-small-cell lung cancer; NGS, next-generation se-
quencing; PD-L1, programmed death-ligand 1; RT-PCR, reverse transcriptase polymerase chain reaction; TKI, tyrosine kinase inhibitor.

therefore, may not meet the threshold requirements for testing, likely leading to inadequate testing results.23 We
specific testing platforms24 (optimal tumor content is 10% applied a 38% overestimation rate to the 14% of tissues
for PCR-based tests, 20% for NGS-based tests, 30% for having , 20% tumor content to the share tested by NGS
Sanger sequencing-based tests). We calculated that 38% and Sanger and generated a potential patient loss rate of
of the 14% of tissue samples having , 20% tumor cell 1.7%. Practice gap 3 led to 14 of 798 patients lost. The
content were overestimated to have . 20% and were thus remaining 784 patients advance on the precision oncology
erroneously deemed appropriate to proceed for molecular care pathway.

Step 1 Step 2 Step 3 Step 4 Step 5 Step 6 Step 7

Physicians
not ordering
Tissue testing
TAT – result Targeted treatment not
sufficiency
Lack of No results not reported selected despite positive
Potential
Tissue and/or awareness Premature reported Test within test result
practice gaps Insufficient Tumor load
liquid biopsy not of current treatment (QNS/TNP/ performance treatment
tumor overestimation guidelines decision
performed initiation inconclusive /sensitivity
for NSCLC rates) window Report indicates
Initial alternative/no therapy
testing
biopsy Rebiopsy
Insurance
challenges
Medicare CMS Claims data (parts A,
Medicare Medicare claims data Medicare claims
Medicare claims data claims data Medicare B and D)
Data sources claims data Published journals Real-time laboratory data
Published journals Published claims data Real-time laboratory data
SEER data Real-time laboratory data Published journals
journals Published journals

Patients
1000 934 798 784 642 524 503
available

% of patients lost 6.6 4.0 0.97 9.6 0 1.7 17.5 0.6 14.5 3.9 4 29.2

Patients
934 798 784 642 524 503 356
advancing

Total patients
lost 66 136 14 142 118 21 147

FIG 2. The precision oncology care pathway: Overall impact of clinical practice gaps on the loss of eligible patients to the delivery of personalized advanced
NSCLC care. CMS, Centers for Medicare & Medicaid Services; NSCLC, non–small-cell lung cancer; TAT, turnaround time; TNP, test not performed; QNS,
quantity not sufficient.

JCO Precision Oncology 5

Downloaded from ascopubs.org by 125.165.106.29 on February 23, 2023 from 125.165.106.029

Copyright © 2023 American Society of Clinical Oncology. All rights reserved.
 Sadik et al

Clinical Practice Gap 4: Appropriate Testing Was Not not consider molecular testing insights. We analyzed the
Ordered or Treatment Began Before Testing Was Ordered laboratory TAT capabilities of the top 30 aNSCLC US
Of the 35,556 patients with aNSCLC who had either a tissue laboratories (representing 68.2% of aNSCLC market share)
or liquid biopsy, 82.5% had biomarker testing ordered for each of the following methodologies: fluorescence in situ
while 17.5% did not have any biomarker testing ordered. hybridization, IHC, NGS, and molecular (Sanger se-
Many potential barriers may have affected test ordering for quencing, ddPCR, and reverse transcriptase PCR). Dif-
these patients. On the basis of an International Association fering testing platforms are associated with different TATs,
for the Study of Lung Cancer (IASLC) survey (United States but NGS testing was associated with the highest average
plus Canada data),36 we estimate that of the 17.5% of TAT. We determined a 13.1% rate of problematic TAT by
patients with no tests ordered, 7.4% were lost because of NGS (. 14 days as noted in guideline from the College of
cost concerns, 2.2% were lost because of test accessibility, American Pathologists, IASLC, and the Association for
1.7% were lost because of a lack of awareness of testing Molecular Pathology)25 for patients receiving results and
options, 1.5% were lost because of low confidence in test applied this to the NGS-tested share for aNSCLC to estimate
value by the ordering physician or the patient, and 4.7% the number of patients who had treatment initiation before
were lost because of other reasons. the return of testing results (Data Supplement).25

On the basis of data from Medicare claims, 3.2% of patients Practice gap 6 has led to 21 of 524 patients lost. The
received treatment before the ordering of biomarker testing. remaining 503 patients advance on the precision oncology
Of the patients who were quickly treated with immuno- care pathway.
therapy, chemotherapy, or antivascular endothelial growth
Clinical Practice Gap 7: Targeted Treatment Was Not
factor treatment, we estimate that 21.2% may have had an
Selected Despite Positive Test Results
actionable biomarker, causing an additional 0.6% of eligible
patients to be lost on the precision oncology care pathway. Of the 27,186 patients who received biomarker testing and
received a timely result, we estimate that 29.2% did not
Practice gap 4 has led to 142 of 784 patients lost. The
receive the appropriate targeted treatment on the basis of
remaining 642 patients advance on the precision oncology
their test results. On the basis of claims data, we deter-
care pathway.
mined that 18.5% of patients received no treatment.
Clinical Practice Gap 5: Biomarker Testing Provided Reasons for a patient to not receive a treatment are pre-
Inconclusive or FN Results sumed and may include patient death before the initiation
Of the patients who received biomarker testing, we estimate of treatment, patient transferred to hospice care, or patient
that 14.5% had an uninformative/inconclusive result be- or caregiver electing to forego treatment. 81.5% of patients
cause of several factors. On the basis of laboratory-level received various treatments, including 9.1% given che-
data, we show that of the 14.5% of uninformative results, motherapy only, 14.3% given chemotherapy and immu-
7.5% is due to technical failure leading to the test not being notherapy, 40.8% given immunotherapy only, 16.6% given
performed, 5.8% is due to sample quantity/quality not targeted therapy, and 0.7% given other treatments.
sufficient for testing that had not been detected during Unmatched or inappropriate treatment after detection of an
preanalytic processing, and 1.1% is due to inconclusive actionable mutation may occur for various reasons. We
data (Data Supplement). found that 14.3% of patients whose tumors were deter-
We applied test method-dependent sensitivity rates to mined to have actionable results that determine targeted
patients who received a single molecular test to show the drug eligibility, such as for tyrosine kinase inhibitors, did not
average level of FN rates for individual tests and to estimate receive the indicated treatment. An additional 11.6% of
the number of eligible patients lost because of FN rates patients with actionable results on the basis of IHC testing,
(Data Supplement). Of the patients who received testing for such as to inform the use of antiprogrammed death-ligand
actionable biomarker(s) and had a result reported, we
1 therapies, did not receive the appropriate immunother-
estimate that 3.9% received a FN result. False positives are
apy. A total of 3.3% of patients were estimated to have
addressed as part of practice gap 7.
incorrect test results (false positives) (Data Supplement).
Practice gap 5 has led to 118 of 642 patients lost. The
remaining 524 patients advance on the precision oncology The reasons for failure to act on positive test results were not
care pathway. determined in this analysis but can depend on various
barriers related to receiving targeted therapies, such as
Clinical Practice Gap 6: As a Result of Turnaround Time reporting issues (errors in reports, outdated clinical infor-
Delays, Treatment Was Initiated Without Consideration of mation, and missing drugs); lack of US Food and Drug
Test Results Administration–approved indication (physician unclear or
Of the 29,227 patients receiving biomarker testing with unwilling to use off-label); and lagging awareness of tar-
reported results, we estimate that 4% experienced labo- geted treatment options and/or guidance, social determi-
ratory TAT delays that led to treatment decisions that did nants of health access/disparities (therapy accessibility

6 © 2022 by American Society of Clinical Oncology

Downloaded from ascopubs.org by 125.165.106.29 on February 23, 2023 from 125.165.106.029

Copyright © 2023 American Society of Clinical Oncology. All rights reserved.
 Clinical Practice Gaps Affecting Precision Oncology

constraints); therapy cost/insurance coverage concerns; or showed variable testing rates for different cancers3,5 and rates
comorbidities/contraindications (Fig 3). of patients who test positive for actionable biomarkers but do
Practice gap 7 has led to 147 of 503 patients lost. The not receive targeted therapies.8,9 In our evaluation of practice
remaining 356 patients advance on the precision oncology gap 1, the number of lost patients who never received an initial
care pathway. biopsy corroborates with NSCLC data from the SEER Program
on cancer statistics.14 Overestimation of tumor content load
Results Summary
(evaluated in practice gap 3) can lead to FN results for Sanger
Overall, we estimate that for every 1,000 patients with newly sequencing- and NGS-based tests. However, this is not the
diagnosed aNSCLC who are potentially eligible for targeted only factor driving FN results. In our evaluation of practice gap
therapy, a cumulative tally of 497 are lost to factors associated 5, we focused mainly on FN rates driven by causes other than
with getting a biomarker test while 147 biomarker-positive tumor load overestimation and expanded our analysis to all
patients are not prescribed the appropriate targeted therapy technologies, leaving minimal overlap between these two
(Fig 4). analyses. It is important to keep in mind that although the
Thus, 644 of every 1,000 patients with newly diagnosed different gap analyses are shown as discrete steps, they are all
aNSCLC (64.4%) are not benefiting from precision on- connected and influence the overall delivery of care. Here, we
cology care options appropriate for their diseases and will combine data reflecting all of the clinical practice gaps to
likely have suboptimal outcomes. clearly show the magnitude of the issues and to highlight that
DISCUSSION more patients than previously expected are actually not re-
ceiving the most appropriate treatment.
We found that only 356 of 1,000 (approximately 36% of)
eligible patients with aNSCLC are benefitting from precision Our estimations are subject to various limitations. We have
oncology treatments appropriate for their condition. Our re- only analyzed patients with procedures and treatments that
sults show that 497 of 1,000 (approximately 50% of) patients are included within claims data cross-referenced to labo-
are lost along the precision oncology pathway because of ratory data, so patients who are enrolled in clinical trials who
preanalytic and analytical practice gaps related to biospeci- receive procedures that are not represented in a claim
men processing or diagnostic test ordering, performance, and would not be included. Although we used practice-based
reporting. Of the 503 of 1,000 patients who do get biomarker data from the Diaceutics repository whenever possible, we
testing and have reported actionable mutations, approxi- relied on published data to supplement the 2019 repository
mately 147 (29.2%) are lost because of post-testing practice data, sometimes from sources that include data from earlier
gaps. Although the relative impact of each clinical practice time periods. Although we have used the most recent and
gap will vary across health care delivery institutions, our es- relevant data available to determine the current impact of
timates provide baseline indications for the overall US health clinical practice gaps, some of the older published data
care system. The findings convey a sense of the magnitude of may not reflect current evolving practice standards.
the challenge associated with each practice gap, which can The analysis of practice gap 4, biomarker test ordering, was
help catalyze and inform strategies to address them. based on claims data from the Diaceutics Data Repository,
Our results generally concur with several previously published but evaluation of factors that can lead to a test not being
studies examining biosample processing efficiency and the ordered was based on US and Canada IASLC data. However,
delivery of personalized medicine, including studies that even with survey data providing reasons why treating

Primary reasons for failing to act on positive

predictive biomarker test results
Patients with positive predictive
biomarker test results who did Reporting issues (errors in report,
not receive therapy outdated clinical information, missing
FIG 3. Barriers to receiving
drugs, and misinterpretation)
biomarker-based therapy
for patients with actionable Lack of FDA-approved indications
Failure to act on positive biomarker
mutations. FDA, US Food testing results
Lack of awareness of targeted treatment
and Drug Administration; option and/or guidance
TKI (14.3%)
FP, false positive; ICI, im- ICI (11.6%)
mune checkpoint inhibitor; Social determinants of health
TKI, tyrosine kinase inhibitor. access/disparities
Incorrect test results Therapy cost/insurance coverage
FP rate (3.3%) concerns

Comorbidities/contraindications

JCO Precision Oncology 7

Downloaded from ascopubs.org by 125.165.106.29 on February 23, 2023 from 125.165.106.029

Copyright © 2023 American Society of Clinical Oncology. All rights reserved.
 Sadik et al

Practice gap 1: Practice gap 2: Practice gap 3: Practice gap 4: Practice gap 5: Practice gap 6: Practice gap 7:
biopsy referral biospecimen biospecimen biomarker test biomarker test result treatment
collection evaluation/pathology ordering testing reporting decision
performance

Patients with newly Potentially Potentially Potentially Potentially Potentially

diagnosed aNSCLC eligible eligible Potentially
eligible eligible eligible Patients
potentially eligible patients patients eligible patients
patients patients patients treated
for targeted therapy remaining remaining remaining
remaining remaining remaining (n = 356)
(n = 1,000) (n = 934) (n = 798) (n = 642)
(n = 784) (n = 524) (n = 503)

6.6% 14.6% 1.7% 18.1% 18.4% 4.0% 29.2%

patients patients patients patients patients patients patients
lost: lost: lost: lost: lost: lost: lost:
66/1000 136/934 14/798 142/784 118/642 21/524 147/503

FIG 4. Impact of clinical practice gaps on the delivery of precision oncology for aNSCLC. aNSCLC, advanced non-small cell lung cancer.

physicians did not order a test for their patients, the true driver options and immune checkpoint inhibitors that can be used to
or combination of drivers is complex and remains subject to inform immunotherapeutic options. However, we have sep-
estimation. When evaluating test performance rates, we cal- arated out IHC-based testing for immune checkpoint inhibi-
culated average positive testing rates using laboratory data tors from genetic testing to help show the relative impact on
within the Diaceutics Data Repository and data from the the clinical practice gaps of each of these biomarker types.
American Association for Cancer Research’s Genomics Evi-
In conclusion, this analysis leverages claims and laboratory
dence Neoplasia Information Exchange project (an interna-
data to provide real-world evidence demonstrating the im-
tional pan-cancer registry of real-world data)15 weighted by the
pact of various clinical practice gaps on the delivery of
number of patients tested by laboratories represented in both
precision oncology care. Only approximately 36% of patients
databases. Although this strategy is meant to allow for the
with aNSCLC are benefiting from precision oncology, indi-
largest possible statistical sampling to drive the determination
of average positivity rates, actual rates may differ depending cating a significant clinical impact deficit. This finding
on the patient cohort. When evaluating the number of patients warrants further investigation, investment, and action. De-
lost to precision oncology because of TAT, we based our cision makers would do well to consider the impact of each
evaluation on laboratory TAT only, while, in fact, total TAT can practice gap when prioritizing and developing process and
be longer because of multiple reasons and can affect more practice standards designed to improve the delivery of
patients. clinical care. Although this study highlights the need for
efforts to address all practice gaps and to develop precision
Although causes of practice gaps related to test ordering,
processing, and performance are relatively clear, the causes oncology implementation improvement strategies, it repre-
of downstream gaps are less clear. Patients who receive sents a snapshot in time using data from 2019. These results
suboptimal treatment because of misinterpretation of test should be updated over time to gauge progress and to show
results are not differentiated from patients who had appro- the impact of improved delivery efforts. Although these
priate interpretation of test results but still did not get treatment findings are made in aNSCLC, they are likely reflective of
matched to their results. Physicians may fail to act on positive similar gaps in other cancer types, indicating a need for
biomarker test results because of a number of factors (Fig 3). further investigation and action across oncology practices.
The relative impact of each of these causes is not clear. Addressing practice gaps can lead to improved clinical care
Nonetheless, these issues have been identified as barriers to associated with a precision oncology approach. Attention to
the delivery of personalized treatment,3,37,38 and efforts to practice gaps may also help to decrease health care costs
address each of these challenges should incentivize or im- through enhanced systemic efficiency and potentially re-
prove the use of biomarker test results to deliver appropriate duced downstream spending on hospitalizations and
targeted therapies and reduce the overall impact of this health resource expenditures necessitated by suboptimal
practice gap. earlier care. An increased understanding of the impact of
Our analysis considers precision oncology biomarkers in- practice gaps can thus inform strategies to deliver more
cluding genomic variants that can inform targeted therapeutic fully on the promise of personalized medicine.

8 © 2022 by American Society of Clinical Oncology

Downloaded from ascopubs.org by 125.165.106.29 on February 23, 2023 from 125.165.106.029

Copyright © 2023 American Society of Clinical Oncology. All rights reserved.
 Clinical Practice Gaps Affecting Precision Oncology

AFFILIATIONS Helen Sadik

1
Diaceutics, Belfast, United Kingdom Employment: Diaceutics
2
Personalized Medicine Coalition, Washington, DC Stock and Other Ownership Interests: Diaceutics
3
Reservoir Communications Group, Washington, DC Daryl Pritchard
Honoraria: Xcenda, Genentech
CORRESPONDING AUTHOR Research Funding: Thermo Fisher, AstraZeneca (Inst)
Daryl Pritchard, PhD, Personalized Medicine Coalition, 1710 Rhode Travel, Accommodations, Expenses: Genentech
Island Ave, NW, Washington, DC 20036; e-mail: Derry-Mae Keeling
[email protected]. Stock and Other Ownership Interests: GlaxoSmithKline

Frank Policht
SUPPORT
Employment: PATHAI
Supported in part by the Personalized Medicine Coalition, a nonprofit
Stock and Other Ownership Interests: Abbott Laboratories, AbbVie
501c3 organization dedicated to the advancement of personalized
medicine. H.S.: Thermo Fisher Scientific, AstraZeneca, Eli Lilly and Peter Riccelli
Company, Blueprint Medicines, Oncocyte. Stock and Other Ownership Interests: Diaceutics, HTG Molecular
Diagnostics
DATA SHARING STATEMENT Jeff Schreier
Raw data were generated at Diaceutics PLC (Belfast, United Kingdom). Employment: Diaceutics Inc
Derived data supporting the findings of this study are available from the Stock and Other Ownership Interests: Diaceutics Inc
lead author (H.S. [email protected]) on request.
Susanne Munksted
Employment: Diaceutics, Agilent
AUTHOR CONTRIBUTIONS Leadership: Diaceutics
Conception and design: Helen Sadik, Daryl Pritchard, Frank Policht, Peter Stock and Other Ownership Interests: Diaceutics
Riccelli, Gretta Stone, Jeff Schreier, Susanne Munksted
Financial support: Daryl Pritchard No other potential conflicts of interest were reported.
Administrative support: Daryl Pritchard, Kira Finkel, Jeff Schreier,
Susanne Munksted ACKNOWLEDGMENT
Provision of study materials or patients: Susanne Munksted Authors Frank Policht and Peter Riccelli contributed to this study while
Collection and assembly of data: Helen Sadik, Derry-Mae Keeling, Peter employed by Diaceutics but are now affiliated with PathAI and HTG
Riccelli, Kira Finkel, Susanne Munksted Molecular, respectively. The authors acknowledge the important input
Data analysis and interpretation: Helen Sadik, Daryl Pritchard, Derry-Mae from members of our project advisory committee including Bonnie
Keeling, Frank Policht, Peter Riccelli, Jeff Schreier, Susanne Munksted Addario (GO2 Foundation for Lung Cancer), Ronnie Andrews (Oncocyte),
Manuscript writing: All authors Francesca Angeletti (Amgen), Aarti Balch (Amgen), Mathew Beer
Final approval of manuscript: All authors (Thermo Fisher Scientific), Aaron Bowles (Blueprint Medicines), Amy
Accountable for all aspects of the work: All authors Carroll (Thermo Fisher Scientific), Cheryl Daniels (AstraZeneca), James
Devlin (Quest Diagnostics), Rob Dumanois (Thermo Fisher Scientific),
AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF Laura Eldon (Flatiron Health), Andrea Ferris (LUNGevity Foundation),
Yuri Fesko (Quest Diagnostics), Jennifer Floyd (Amgen), Lisa Giuroiu
INTEREST
(Amgen), Mark Hiatt (Guardant Health), Aubrey Kelly (RabbleHealth),
The following represents disclosure information provided by authors of
Yonghong Li (Quest Diagnostics), Nikki Martin (LUNGevity Foundation),
this manuscript. All relationships are considered compensated unless
Jonathan Meek (Novartis), Casey O’Neil (Genentech), Omar Perez
otherwise noted. Relationships are self-held unless noted. I = Immediate
(AstraZeneca), Elissa Quinn (Blueprint Medicines), Julie Ramage
Family Member, Inst = My Institution. Relationships may not relate to the
(AstraZeneca), Kristen Santiago (LUNGevity Foundation), Timothy
subject matter of this manuscript. For more information about ASCO’s
Showalter (Flatiron Health), Davida Silverman (Genentech), Nino Sireci
conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.
(Eli Lilly and Company), Koji Sonoda (RabbleHealth), and Emma
org/po/author-center.
Van Hook (Eli Lilly and Company). The authors wish to provide special
Open Payments is a public database containing information reported by
recognition to our project physician advisors, Lincoln Nadauld
companies about payments made to US-licensed physicians (Open
(Intermountain Health), Alex Spira (US Oncology Network), and Apostolia
Payments).
Tsimberidou (MD Anderson Cancer Center). Authors also acknowledge
Christopher Wells (Personalized Medicine Coalition) for providing
technical editing and helpful contributions to report development.

REFERENCES
1. Personalized Medicine at FDA: A Progress & Outlook Report. https://www.personalizedmedicinecoalition.org/Userfiles/PMC-Corporate/file/Personalized_
Medicine_at_FDA_The_Scope_Significance_of_Progress_in_2021.pdf
2. The Evolution of Biomarker Use in Clinical Trials for Cancer Treatments: Key Findings and Implications. https://www.personalizedmedicinecoalition.org/
Userfiles/PMC-Corporate/file/The_Evolution_of_Biomarker_Use_in_Clinical_Trials_for_Cancer_Treatments.pdf
3. IQVIA: Optimizing Oncology Care Through Biomarker Adoption-Barriers and Solutions. 2019. https://www.iqvia.com/insights/the-iqvia-institute/reports/
optimizing-oncology-care-through-biomarker-adoption
4. Huang RSP, Severson E, Haberberger J, et al: Landscape of biomarkers in non-small cell lung cancer using comprehensive genomic profiling and PD-L1
immunohistochemistry. Pathol Oncol Res 27:592997, 2021
5. Robert NJ, Nwokeji ED, Espirito JL, et al: Biomarker tissue journey among patients (Pts) with untreated metastatic non-small cell lung cancer (Mnsclc) in the
U.S. Oncology Network community practices. J Clin Oncol 39, 2021 (15_suppl; abstr 9004)

JCO Precision Oncology 9

Downloaded from ascopubs.org by 125.165.106.29 on February 23, 2023 from 125.165.106.029

Copyright © 2023 American Society of Clinical Oncology. All rights reserved.
 Sadik et al

6. Diaceutics Finds 156K Cancer Patients Could Miss Out on Life-Saving Drugs Every Year in US and Europe. Diaceutics. 2017. https://www.diaceutics.com/
diaceutics-finds-156k-cancer-patients-could-miss-out-on-life-saving-drugs-every-year-in-us-and-europe/
7. Clark J, Smart D, Keeling P: Patients lost to precision treatment due to suboptimal testing: US and EU real world evidence. Presented at European Alliance for
Personalised Medicine Congress (EAPM), Belfast City, North Ireland, November 27-30, 2017. https://www.diaceutics.com/wp-content/uploads/2019/05/
eapm-2017.pdf
8. Steuten, L, Goulart, B, Meropol, NJ, et al: Cost effectiveness of multigene panel sequencing for patients with advanced non-small cell lung cancer. JCO Clin
Cancer Inform 3:1-10, 2019
9. Presley CJ, Tang D, Soulos PR, et al: Association of broad-based genomic sequencing with survival among patients with advanced non-small cell lung cancer in
the community oncology setting. JAMA 320:469-477, 2018
10. Bunn PA Jr, Aisner DL: Broad-based molecular testing for lung cancer: Precisely the time for precision. JAMA 320:445-446, 2018
11. Vashistha V, Armstrong J, Winski D, et al: Barriers to prescribing targeted therapies for patients with NSCLC with highly actionable gene variants in the Veterans
Affairs National Precision Oncology Program. JCO Oncol Pract 17:e1012-e1020, 2021
12. Goulart BHL, Chennupati S, Fedorenko CR, et al: Access to tyrosine kinase inhibitors and survival in patients with advanced EGFR+ and ALK+ positive non-
small-cell lung cancer treated in the real-world. Clin Lung Cancer 22:e723-e733, 2021
13. Introducing PxRx - The Diagnostic Network. Diaceutics. https://www.diaceutics.com/
14. Surveillance, Epidemiology, and End Results (SEER) Program, National Cancer Institute, DCCPS, Surveillance Research Program, USA data released 2017.
https://seer.cancer.gov/statfacts/html/lungb.html
15. The AACR Project GENIE Consortium: AACR project GENIE: Powering precision medicine through an international consortium, Cancer Discov 7:818-831, 2017
16. Waterhouse D, Lam J, Betts KA, et al: Real-world outcomes of immunotherapy-based regimens in first-line advanced non-small cell lung cancer. Lung Cancer
156:41-49, 2021
17. Gutierrez ME, Choi K, Lanman RB, et al: Genomic profiling of advanced non-small cell lung cancer in community settings: Gaps and opportunities. Clin Lung
Cancer 18:651-659, 2017
18. Coley SM, Crapanzano JP, Saqi A: FNA, core biopsy, or both for the diagnosis of lung carcinoma: Obtaining sufficient tissue for a specific diagnosis and
molecular testing. Cancer Cytopathol 123:318-326, 2015
19. Aisner DL, Rumery MD, Merrick DT, et al: Do more with less: Tips and techniques for maximizing small biopsy and cytology specimens for molecular and
ancillary testing: The University of Colorado experience. Arch Pathol Lab Med 140:1206-1220, 2016
20. U.S. Food and Drug Administration, Center for Devices and Radiological Health: Cobas® EGFR Mutation Test, PMA P120019 Summary of Safety and
Effectiveness Data, 05/2013. https://www.accessdata.fda.gov/cdrh_docs/pdf12/P120019B.pdf
21. Warth A, Penzel R, Brandt R, et al: Optimized algorithm for Sanger sequencing-based EGFR mutation analyses in NSCLC biopsies. Virchows Arch
460:407-414, 2012
22. U.S. Food and Drug Administration, Center for Devices and Radiological Health: OncomineDx Target Test, PMA P160045 User Guide, 11/2020. https://www.
accessdata.fda.gov/cdrh_docs/pdf16/P160045S019C.pdf
23. Smits AJ, Kummer JA, de Bruin PC, et al: The estimation of tumor cell percentage for molecular testing by pathologists is not accurate. Mod Pathol 27:168-174,
2014
24. Patel NM, Jo H, Eberhard DA, et al: Improved tumor purity metrics in next-generation sequencing for clinical practice: The integrated interpretation of neoplastic
cellularity and sequencing results (IINCaSe) approach. Appl Immunohistochem Mol Morphol 27:764-772, 2019
25. Lindeman NI, Cagle PT, Aisner DL, et al: Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase
inhibitors: Guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular
Pathology. Arch Pathol Lab Med 142:321-346, 2018
26. U.S. Food and Drug Administration, Center for Devices and Radiological Health: FoundationOne CDx, PMA P170019 Technical Information, 12/2017.
https://www.accessdata.fda.gov/cdrh_docs/pdf17/P170019C.pdf
27. U.S. Food and Drug Administration, Center for Devices and Radiological Health: Guardant360® CDx, PMA P200010 Technical Information, 08/2020. https://
www.accessdata.fda.gov/cdrh_docs/pdf20/P200010C.pdf
28. U.S. Food and Drug Administration, Center for Devices and Radiological Health: Therascreen® EGFR RGQ PCR Kit, PMA P120022 Summary of Safety and
Effectiveness Data, 07/2013. https://www.accessdata.fda.gov/cdrh_docs/pdf12/P120022B.pdf
29. Tsiatis AC, Norris-Kirby A, Rich RG, et al: Comparison of Sanger sequencing, pyrosequencing, and melting curve analysis for the detection of KRAS mutations:
Diagnostic and clinical implications. J Mol Diagn 12:425-432, 2010
30. Zeng X, Lian T, Lin J, et al: Whole-exome sequencing improves genetic testing accuracy in pulmonary artery hypertension. Pulm Circ 8:2045894018763682, 2018
31. Harismendy O, Ng PC, Strausberg RL, et al: Evaluation of next generation sequencing platforms for population targeted sequencing studies. Genome Biol
10:R32, 2009
32. Marchetti A, Barberis M, Papotti M, et al: ALK rearrangement testing by FISH analysis in non-small-cell lung cancer patients: Results of the first Italian external
quality assurance scheme. J Thorac Oncol 9:1470-1476, 2014
33. U.S. Food and Drug Administration, Center for Devices and Radiological Health: VENTANA ALK (D5F3) CDx Assay, PMA P140025 Summary of Safety and
Effectiveness Data, 06/2015. https://www.accessdata.fda.gov/cdrh_docs/pdf14/P140025B.pdf
34. U.S. Food and Drug Administration, Center for Devices and Radiological Health: Vysis ALK Break Apart FISH Probe Kit, PMA P110012 Summary of Safety and
Effectiveness Data, 08/2011. https://www.accessdata.fda.gov/cdrh_docs/pdf11/P110012B.pdf
35. U.S. Food and Drug Administration, Center for Devices and Radiological Health: FoundationOne® Liquid CDx, PMA P200016 Technical Information, 11/2020.
https://www.accessdata.fda.gov/cdrh_docs/pdf20/P200016C.pdf
36. Smeltzer MP, Wynes MW, Lantuejoul S, et al: The International Association for the Study of Lung Cancer Global Survey on molecular testing in lung cancer.
J Thorac Oncol 15:1434-1448, 2020
37. Tsimberidou AM, Elkin S, Dumanois R, et al: Clinical and economic value of genetic sequencing for personalized therapy in non-small-cell lung cancer. Clin
Lung Cancer 21:477-481, 2020
38. Chan IS, Ginsburg GS: Personalized medicine: Progress and promise. Annu Rev Genomics Hum Genet 12:217-244, 2011

n n n

10 © 2022 by American Society of Clinical Oncology

Downloaded from ascopubs.org by 125.165.106.29 on February 23, 2023 from 125.165.106.029

Copyright © 2023 American Society of Clinical Oncology. All rights reserved.