JNCI J Natl Cancer Inst (2021) 113(11): djaa144

doi: 10.1093/jnci/djaa144
First published online September 22, 2020
Commentary

The Impact of COVID-19 on Clinical Trial Execution at the Dana-Farber

Cancer Institute
Sara M. Tolaney, MD, MPH,1 Christine A. Lydon, BA,1 Tianyu Li, MS,2 Jiale Dai, PhD, RPH,3
Andrea Standring, PharmD,3 Kristen A. Legor , JD, RN, OCN,4 Caryn M. Caparrotta, BSN, RN, OCN,4
5
Matthew P. Schenker , MD, Daniel I. Glazer , MD,5 Nabihah Tayob, PhD,2 Steven G. DuBois, MD,6
Jeffrey A. Meyerhardt , MD,1 Mary-Ellen Taplin, MD,1 Bruce E. Johnson, MD1,*

COMMENTARY
1
Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA; 2Division of Biostatistics, Department of Data Sciences, Dana-Farber Cancer
Institute, Boston, MA, USA; 3Pharmacy and Clinical Support, Dana-Farber Cancer Institute, Boston, MA, USA; 4Nursing and Patient Care Services, Dana-Farber Cancer
Institute, Boston, MA, USA; 5Department of Radiology, Brigham and Women’s Hospital, Boston, MA, USA and 6Pediatric Oncology, Dana-Farber Cancer Institute,
Boston, MA, USA

*Correspondence to: Bruce E. Johnson, MD, 450 Brookline Ave, Boston, MA 02215, USA (e-mail: bruce_johnson@dfci.harvard.edu).

Abstract
Interventions designed to limit the spread of coronavirus disease 2019 (COVID-19) are having profound effects on the delivery
of health care, but data showing the impact on oncology clinical trial enrollment, treatment, and monitoring are limited. We
prospectively tracked relevant data from oncology clinical trials at Dana-Farber Cancer Institute from January 1, 2018, to June
30, 2020, including the number of open trials, new patient enrollments, in-person and virtual patient visits, dispensed investi-
gational infusions, dispensed or shipped oral investigational agents, research biopsies, and blood samples. We ascertained
why patients came off trials and determined on-site clinical research staffing levels. We used 2-sided Wilcoxon rank sum
tests to assess the statistical significance of the reported changes. Nearly all patients on interventional treatment trials were
maintained, and new enrollments continued at just under one-half the prepandemic rate. The median number of investiga-
tional prescriptions shipped to patients increased from 0 to 74 (range ¼ 22-107) per week from March to June 2020. The me-
dian number of telemedicine appointments increased from 0 to 107 (range ¼ 33-267) per week from March to June 2020.
Research biopsies and blood collections decreased dramatically after Dana-Farber Cancer Institute implemented COVID-19–
related policies in March 2020. The number of research nurses and clinical research coordinators on site also decreased after
March 2020. Substantial changes were required to safely continue clinical research during the pandemic, yet we observed no
increases in serious adverse events or major violations related to drug dosing. Lessons learned from adapting research practi-
ces during COVID-19 can inform industry sponsors and governmental agencies to consider altering practices to increase oper-
ational efficiency and convenience for patients.

A previously unknown respiratory illness emerged in Wuhan, have impacted travel, public gatherings, and access to medical
China, in December 2019. The causative agent was identified as care in the inpatient and outpatient settings, including elective
the novel severe acute respiratory syndrome coronavirus 2 procedures and nonemergent clinical services (6). Oncology care
(SARS-CoV-2), and the associated disease was named coronavi- is commonly critical, and most of the care is administered in
rus disease 2019 (COVID-19) (1–3). The first verified COVID-19 the outpatient setting. This has led members of the oncology
patient in the United States became ill in January 2020 in community to provide guidance for managing the treatment of
Washington State (4). The virus has now spread rapidly cancer patients during this pandemic (7,8). Clinical trials are es-
throughout the United States and caused more than 5.6 million sential for advancing cancer treatment and also provide access
cases and nearly 176 000 deaths by late August of 2020 (5). to novel and potentially effective treatments for patients. The
The spread of this deadly virus has led to substantial public policies enacted around the United States to reduce the
changes in public policies throughout the United States that spread of COVID-19 have prompted the National Cancer

Received: June 3, 2020; Revised: August 27, 2020; Accepted: September 4, 2020
© The Author(s) 2020. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com

1453
 1454 | JNCI J Natl Cancer Inst, 2021, Vol. 113, No. 11

Institute (9) and the Food and Drug Administration (10) to issue prospectively monitored and enumerated. The number of re-
guidelines around the conduct of clinical trials. search blood samples collected was retrieved by monitoring the
Dana-Farber Cancer Institute (DFCI) and Brigham Health kits used for drawing the research blood. The number of in-
(BH) are located in Massachusetts, which had the 12th highest person patient visits and virtual visits by week was retrieved
number of COVID-19 cases and the 4th highest number of from the electronic medical records. The number of investiga-
deaths in the United States by early August 2020 (5). As the tional infusions (from January 1, 2018, to June 30, 2020), oral
threat emerged, the Commonwealth of Massachusetts enacted agents dispensed in the outpatient pharmacy, and oral agents
a series of public health policies to protect its citizens and visi- shipped to patients were prospectively monitored and retrieved
tors. Governor Charles Baker issued a state of emergency to re- from the electronic investigational drug accountability system
spond to COVID-19 on March 10, 2020, when there were 91 (nCoup, Fremont, CA). Research management provided research
known cases of COVID-19 in the state (11). DFCI and BH followed nursing and clinical research coordinator (CRC) staffing levels
the policies set forth by the Commonwealth of Massachusetts, from March 16, 2020, to June 30, 2020, the number of trials that
including a stay-at-home advisory on March 23, restrictions on held accrual due to COVID-19, and the number of patients who
gatherings of 10 or more people, and other policies that are af- came off trials due to COVID-19.
fecting the methods and the ability to deliver health care to our
patients. In addition, DFCI and BH had to ensure that there were
adequate resources to care for the anticipated surge in COVID- Statistical Analysis
19 admitted inpatients in April of 2020 (12,13). On May 18, 2020,
The statistics team used various statistical methods and models
Governor Baker instituted phase I of the stepwise reopening of
to assess the clinical research changes over time. The Wilcoxon
businesses and other organizations in Massachusetts. Phase II,
COMMENTARY

rank sum test (2-sided) was used to evaluate the time effect on
step 1 was subsequently initiated on June 1, 2020, followed by
the following factors: the number of patients enrolled to inter-
phase II, step 2 on June 19. In response, DFCI and BH instituted a
ventional treatment studies; the number of patients enrolled to
phased recovery plan beginning on May 18, 2020 (11–13).
interventional, nontreatment studies; the number of patients
Although the cancer community has provided guidance for
enrolled to noninterventional studies; the number of interven-
managing patients with cancer and federal agencies have is-
tional treatment trials; the number of infusions of investiga-
sued guidelines for clinical trials, there are little objective data
tional agents; the number of biopsies for research purposes; and
to assess the impact of public and institutional policies for
the number of research blood collections associated with thera-
patients who are considering or already enrolled in clinical tri-
peutic trials. The time factor was trinary, with 2018, 2019, and
als. The Office of Clinical Research at DFCI monitors clinical tri-
January-February 2020 collapsed into 1 group; March-May 2020
als by assembling pertinent data and providing feedback to
coded as another group; and June coded as the third group as
study teams. In the months following the enactment of the pub-
DFCI and BH instituted a phased recovery plan coordinated with
lic policies meant to limit the spread of COVID-19, as well as
the State of Massachusetts policies. Linear spline regression
policies to ensure access to acute medical care for patients who
models were fit to assess the following associations: the propor-
contracted the virus, we assessed the ability to carry out inter-
tion of investigational agents dispensed weekly by mail from
ventional treatment clinical trials in this challenging
January to June 2020; the proportion of patients (adult or pediat-
environment.
ric) evaluated by telemedicine appointments over the weeks
from March 16 to June 28, 2020; the number of missed appoint-
ments either in-person or via telemedicine over the weeks from
Methods March 16 to June 28, 2020; the proportion of research nurses
working remotely over the weeks from March 16 to June 28,
Metrics and Data Collection 2020; and the number of CRCs working on-site over the weeks
from March 16 to June 28, 2020. The week of May 31, 2020, was
The Office of Clinical Research has prospectively tracked met-
picked as the spline knot for all the linear spline models. We
rics for clinical trials from 2015 to the present. We retrieved the
wanted to examine whether the recovery policy issued on May
dates and specifics of the COVID-19–related institutional poli-
28, 2020, at DFCI had any effect on the clinical practices previ-
cies that affected clinical care and clinical trials at DFCI from
ously described. The statistics team performed all analyses us-
online staff communications. The objective information
ing SAS 9.4. P values less than .05 were considered statistically
deemed pertinent to the changes in clinical trial execution was
significant.
identified by the chief clinical research officer, deputy chief clin-
ical research officer, the chair of the Executive Committee on
Clinical Research, the associate chief nurse for clinical research,
Results
the director of research nursing, and the administrative director
for the Office of Clinical Research. The list was distributed to the The ongoing guidance from the Commonwealth of
study teams, who gathered the relevant information. Massachusetts, the National Cancer Institute, and the Food and
The number of patients accrued to clinical trials by month Drug Administration prompted DFCI to enact progressively re-
was retrieved from January 1, 2018, through June 30, 2020, from strictive guidelines for clinical research operations from March
the OnCore Clinical Trial Management System Enterprise through April 2020 (Box 1). Operational restrictions at BH made
Research system. The information on other parameters was in preparation for the COVID-19 patient surge further affected
gathered by the study teams from January 1, 2020, to June 30, clinical research at DFCI. Specifically, the procurement of
2020, with a few exceptions as noted below. The number of image-guided percutaneous biopsies for research purposes was
studies that were open but were not able to enroll and the num- temporarily limited to only support research that had the po-
ber of patients and reasons for coming off the clinical trials tential to be lifesaving or disease altering in circumstances
were ascertained. The number of research biopsies performed where no alternative clinical treatments were available. These
each month (from January 1, 2018, to June 30, 2020) was policies necessitated adjustments to our staffing models and
 S. M. Tolaney et al. | 1455

Box 1. Policies for carrying out clinical research at Dana-Farber Cancer Institute (DFCI) from March 13 through June 11, 2020

Restrictions
Level Description Effective date
Level 1 Minimal restriction – Clinical research work may continue as usual, but staff should reduce their time March 13, 2020
on-site and work from home when possible.
Level 2 Staffing restriction – Only staff who are necessary to continue clinical research activities should be on- N/Aa
site.
Level 3 Reduced staffing to carry out clinical research – Those needed for direct interaction with patients to March 16, 2020
carry out clinical research needs. Staff should consolidate their roles to:
• Clinicians needed to see protocol patients
• Research nurses and CRCs who typically secure consents and assess patients on studies
• Research pharmacy staff
• Cell manipulation core facility staff
• Laboratory staff that process research samples from patients enrolled in active clinical trials
• Research staff in Office for Human Research Subjects, Research Informatics Operations, Office of
Data Quality, and Clinical Trials Office will largely be working off-site
Level 4 At Dana-Farber: March 20, 2020
• Nontherapeutic research not directly related to patient care will be suspended.
• Research staff will focus on maintaining active research protocols and patients on these studies.

COMMENTARY
• Continue consenting patients to existing therapeutic protocols as resources allow.
• Only those needed for direct interaction with patients to address clinical research needs can be on
site:
• Clinicians needed to see the protocol patients
• The number of research nurses and CRCs will be consolidated to minimize patient contact.
• Research pharmacy staff
• Cell manipulation core facility staff
• Laboratory staff who process research samples from patients enrolled on active clinical trials
Recovery
Stage Description Effective date
Stage 1 At Dana-Farber May 18, 2020
• Preparations underway to progress to stage 2A
Stage 2A At Dana-Farber May 28, 2020
• PFTs available: resting oximetry, 6-minute walks, exercise oximetry
• CRL began accepting all clinical trial–related samples
• No restrictions to collecting biospecimens for therapeutic clinical trials
• CSIR research biopsies for therapeutic trials may be scheduled as capacity allows (regardless of
whether a sample is sent to BWH Pathology)
• Triplicate ECGs allowed when required by protocol
• Minimum number of CRCs needed to collect biospecimens permitted on-site
Stage 2B At Dana-Farber June 11, 2020
• Study patients should be scheduled for on-campus visits whenever possible
• Tissue banking from BWH surgical procedures (non-CSIR) resumed (June 22, 2020)
• Up to 5 CRCs allowed on-site per disease center per day
• 1 RRN for every 10 patients allowed on-site per day

a
Given the rapid changes in local and national guidance, DFCI escalated from level 1 to 3, without an intermediate level 2 reduction in staffing and services. BWH ¼
Brigham and Women’s Hospital; CRC ¼ clinical research coordinator; CRL ¼ clinical research laboratory; CSIR ¼ cross-sectional interventional radiology; ECG ¼
electrocardiogram; N/A ¼ not applicable; PFT ¼ pulmonary function test; RRN ¼ research registered nurse.

delivery of care to patients enrolled and screened for clinical tri- recovery was implemented on May 28, 2020, in which the mini-
als. The Office of Clinical Research worked with DFCI to provide mum number of CRCs needed to collect samples was permitted
guidance to study teams, and we assessed the impact on on site. Stage 2B was initiated on June 11, which allowed for study
patients and compliance by our staff. patients to be seen in person whenever possible, with up to 5
As of March 20, 2020, DFCI was at level 4 research (Box 1). In CRCs per disease center and 1 research nurse for every 10 trial
addition to other changes, level 4 restrictions limited the number patients allowed on site per day (Box 1).
of research nurses and study coordinators who could be on-site, Trials with eligibility criteria and study requirements for on-
all monitoring of trials was required to be conducted remotely, study biopsies, frequent clinic visits, extended pharmacokinetic
and biospecimen collection was restricted due to the reduction in assessments, and other research blood collections were particu-
on-site staff available to collect and process samples. On May 18, larly affected by level 4 restrictions. In addition, sponsors’ deci-
2020, DFCI initiated stage 1 of recovery, which involved preparing sions to pause enrollment also had an impact. For each clinical
to resume seeing patients in person and collecting biospecimens trial, the principal investigator was asked to ensure the resour-
for therapeutic and nontherapeutic trials. Collection of biospeci- ces needed to enroll and safely maintain the participants on the
mens for therapeutic clinical trials resumed when stage 2A of study were available, and the sponsor agreed to adapt the
 1456 | JNCI J Natl Cancer Inst, 2021, Vol. 113, No. 11

A B
P = .007

P = .08

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Jan Feb Mar Apr May June
Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Jan Feb Mar Apr May June
2019 2020
2019 2020

C
P = .007
COMMENTARY

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Jan Feb Mar Apr May June

2019 2020

Figure 1. Number of patients enrolled in trials by month, January 2019 to June 2020. A) Number of patients who enrolled in interventional, therapeutic trials by month,
January 2019 to June 2020. B) Number of patients who enrolled in interventional, nontherapeutic trials by month, January 2019 to June 2020. C) Number of patients who
enrolled in noninterventional, nontherapeutic trials by month, January 2019 to June 2020. Two-sided Wilcoxon rank sum tests were used to compare the number of
patients enrolled across the 3 time periods: 2018-February 2020, March-May 2020, and June 2020.

Table 1. Impact of COVID-19–related policies on trial accrual, number of open trials, infusions of investigational agents, and collection of re-
search biospecimens per month at Dana-Farber Cancer Institute

Median (range)

Affected metric Jan 2018-Feb 2020 Mar 2020-May 2020 June 2020 Pa

Patients enrolled in treatment trials, No.

Interventional, treatment 205 (157-241) 86 (83-140) 138 .007
Interventional, nontreatment 161 (85-311) 103 (65-119) 202 .08
Noninterventional 1870 (1374-2531) 255 (128-1369) 380 .007
Interventional treatment trials, No. 539 (523-591) 552 (548-568) 549 .31
Infusions of investigational agents 1462 (984-1837) 802 (774-996) 842 .008
Biopsies for research purposes 85 (67-110) 1 (0-52) 17 .006
Blood collection for research purposes 1142 (545-1328) 243 (199-812) 776 .02

a
These P values still show a decreasing trend across time in general. None of the increases from March-May to June was statistically significant.

protocol requirements to existing conditions brought on by the largely stable between February and June 2020. Nevertheless,
pandemic. since the COVID-19–related policies were instituted, 49 of the
A median of 205 patients per month (range ¼ 157-241) en- 568 enrolling trials (8.6%) were closed (temporarily or perma-
rolled in medical oncology interventional treatment studies nently) and 154 (27.1%) were put on hold (many of which
from January 2018 to February 2020. The number fell dramati- were put on hold due to restrictions on biospecimen collec-
cally between March and May 2020 to a median of 86 patients tion). Additionally, 41 institutional review board-approved tri-
(range ¼ 83-140) before beginning to recover in June 2020, with als were not activated as of the end of June 2020 due to
138 accrued (Figure 1, A). In general, the trend was decreasing COVID-19 restrictions. The enactment of COVID-19–related
(P ¼ .007). The recovery from March through May to June was policies also resulted in a statistically significant reduction in
not statistically significant (P ¼ .65). Although new enrollments the monthly infusion of investigational agents, collection of
declined, only 7 of the 2295 (0.3%) active patients already on in- biopsies for research purposes, and blood collection for re-
terventional treatment trials came off study, due to either con- search purposes (Table 1).
cerns about contracting COVID-19 or international travel The research pharmacy at DFCI dispensed a median of 272
restrictions. Similar decreasing trends were observed for outpatient oral investigational prescriptions per week (range ¼
patients enrolled in interventional, nontreatment studies 225-324) from January 5, 2020, through March 22, 2020 (Figure 2,
(Figure 1, B; P ¼ .08) and noninterventional, nontreatment stud- A). The guidance from the National Cancer Institute and the
ies (Figure 1, C; P ¼ .007) (Table 1). Food and Drug Administration called for increased flexibility in
As shown in Table 1, the median number of interventional shipping investigational agents directly to patients. In response,
treatment trials accruing patients at DFCI by month remained in March 2020, the research pharmacy, with the assistance of
 S. M. Tolaney et al. | 1457

A B
3/10: State of emergency 5/18: Recovery phase I 3/10: State of emergency 5/18: Recovery phase I

6/1: Phase II, step 1 6/1: Phase II, step 1

6/19: Phase II, step 2 6/19: Phase II, step 2

1/5 1/12 1/19 1/26 2/2 2/9 2/16 2/23 3/1 3/8 3/15 3/22 3/29 4/5 4/12 4/19 4/26 5/3 5/10 5/17 5/24 5/31 6/7 6/14 6/21 6/28 12/29 1/5 1/12 1/19 1/26 2/2 2/9 2/16 2/23 3/1 3/8 3/15 3/22 3/29 4/5 4/12 4/19 4/26 5/3 5/10 5/17 5/24 5/31 6/7 6/14 6/21 6/28

C D
3/10: State of emergency 5/18: Recovery phase I 3/10: State of emergency 5/18: Recovery phase I

6/1: Phase II, step 1 6/1: Phase II, step 1

6/19: Phase II, step 2

6/19: Phase II, step 2

COMMENTARY
12/29 1/5 1/12 1/19 1/26 2/2 2/9 2/16 2/23 3/1 3/8 3/15 3/22 3/29 4/5 4/12 4/19 4/26 5/3 5/10 5/17 5/24 5/31 6/7 6/14 6/21 6/28 12/29 1/5 1/12 1/19 1/26 2/2 2/9 2/16 2/23 3/1 3/8 3/15 3/22 3/29 4/5 4/12 4/19 4/26 5/3 5/10 5/17 5/24 5/31 6/7 6/14 6/21 6/28

Figure 2. Changes in oral investigational agent distribution, appointment setting, and missed appointments from January 2020 to June 2020. A) Number of oral investi-
gational agents dispensed in person vs shipped to patients by week, January 2020 to June 2020. B) Number of adult appointments conducted in-person and using tele-
medicine, January 2020 to June 2020. C) Number of pediatric appointments conducted in-person and using telemedicine, January 2020 to June 2020. D) Number of adult
missed appointments, in-person and via telemedicine, January 2020 to June 2020.

research nursing and CRCs, began a program to mail investiga- telemedicine appointments for pediatric patients either from
tional prescriptions to participants rather than requiring in- March to May or in June (P ¼ .95, P ¼ .75, respectively).
person pick-up at the DFCI outpatient pharmacy. Ten sponsors We also tracked missed appointments for patients on clini-
provided blanket approval to ship investigational oral agents, cal trials each week from January to June 2020. Missed appoint-
and 2 sponsors did not provide this approval. ments spiked during the week of March 15, 2020 (Figure 2, D), as
The total oral investigational agents dispensed ranged from DFCI was rapidly transitioning from level 1 to level 4 research
232 to 276 (median ¼ 251) per week during the 14-week period restrictions (Box 1). As providers began seeing patients through
between March 26 to the week of May 31, 2020. The number of telemedicine appointments, a roughly equal proportion of
mailed investigational agents ranged from 49 to 107 per week patients missed in-person and telemedicine appointments
(median ¼ 85) from March 29 to the week of May 31, 2020. In to- from late April to early June 2020. Starting from the week of
tal, the median number of investigational prescriptions shipped March 15, the number of missed in-person appointments de-
to patients increased from 0 to 74 (range ¼ 22-107) per week creased to the end of May (P ¼ .04) but became stable during the
from March 2020 to June 2020, and mailed investigational recovery (P ¼ .44). We observed a statistically significant in-
agents represented 32.0% of all dispensed investigational agents crease in the number of missed telemedicine appointments
during this time. The number of mailed investigational agents from the week of March 15 to the end of May (P ¼ .003) and then
decreased to a median of 31 per week (range ¼ 22-37) and repre- a statistically significant decrease during the recovery (P < .001)
sented 11.8% of all dispensed investigational agents from the (Figure 2, D).
week of June 7 to June 28, 2020 (Figure 2, A). We observed a sta- DFCI has 95 research nurses in both medical and pediatric
tistically significant decrease in the percentage of mailed inves- oncology. Before March 16, 2020, no research nurses worked re-
tigational agents from March 26 to the week of May 31, 2020 motely. After the implementation of COVID-19 policies, the
(P ¼ .003); the decreasing trend did not statistically significantly number of research nurses working on-site in the clinic ranged
alter after June (P ¼ .22). from 17 to 36 per week (median ¼ 20) between the weeks of
As part of the COVID-19 policies, providers at DFCI began see- March 16 and May 31, 2020. The number of research nurses
ing many new and established patients through telemedicine working remotely ranged from 43 to 65 per week (median ¼ 59)
appointments during the week of March 16, 2020. The median during this time. In June, the number of research nurses work-
number of telemedicine appointments increased from 0 to 107 ing on-site in the clinic ranged from 24 to 31 per week; the num-
(range ¼ 33-267) per week from March 2020 to June 2020. The ber of research nurses working remotely ranged from 51 to 58
number of adult telemedicine research–related appointments per week during this time. There was a statistically significant
peaked at 267 during the week of April 12 (Figure 2, B). The per- increase in the percentage of nurses working remotely from
centage of telemedicine appointments for adults from March to March to the week of May 31, 2020 (P < .001), then a statistically
the end of May was stable (P ¼ .30), but we report a statistically significant decrease during the recovery (P < .001).
significant decrease beginning in the month of June (P ¼ .02). The In January 2020, 165 CRCs worked on-site at DFCI. Between
number of pediatric telemedicine research–related appointments March 16 and May 31, the number of CRCs on-site per day varied
peaked at 11 during the weeks of April 12 and May 3 (Figure 2, C). between 20 and 41, with a median of 28 on-site each day.
There was no statistically significant change in the percentage of Between June 1 and June 30, the number of CRCs on-site per day
 1458 | JNCI J Natl Cancer Inst, 2021, Vol. 113, No. 11

varied between 39 and 59, with a median of 51 on-site each day. scheduling of study patients for on-campus visits and collection
The number of CRCs on-site per day was stable from March to of biospecimens for therapeutic and nontherapeutic trials.
the end of May (P ¼ .57), but there was a statistically significant It is important to point out several limitations of this pro-
increase during the recovery (P < .001). Most disease centers spective study. Before the institution of the COVID-19–related
maintained a rotating schedule for CRCs. policies, the research pharmacy did not ship oral medications
To assess the impact of COVID-19–related policies on patient directly to patients unless it was an emergency situation and
compliance to protocol therapy, we recorded the number of pro- approved by the sponsor and principal investigator. However,
tocol violations related and not related to dosing of protocol the precise number of shipped medications was not centrally
therapy by month. In January and February of 2020, there was a tracked; this means that the baseline number of investigational
median of 15 dosing-related protocol violations among a me- agents shipped to patients before the week of March 29, 2020, is
dian of 2424 patients on treatment (3.34-8.98  10–3 violations not clearly known but was very rare. Second, the number of
per patient on treatment) and 54 nondosing-related protocol CRCs and research nurses on-site was obtained by surveying
violations (2.13-2.29  10–2 violations per patient on treatment). clinical research managers in each disease center and nursing
From March to May 2020, there was a median of 10 dosing- leadership in medical and pediatric oncology. However, it is
related protocol violations among a median of 2347 patients on possible that not all schedule changes were captured by the
treatment (range ¼ 3.42-5.53  10–3 violations per patient on managers’ records.
treatment) and 24 nondosing-related protocol violations (range It is important to consider that the experience at DFCI might
¼ 9.80  10–3 to 1.63  10–2 violations per patient on treatment). not reflect the situation at all cancer centers around the coun-
In June 2020, there was 1 dosing-related protocol violation and try, because policies and resources differ by center. A recent
15 nondosing-related protocol violations among 2391 patients American Society of Clinical Oncology survey of 32 cancer cen-
COMMENTARY

on treatment. ters (14 academic centers and 18 community programs)

In addition, we assessed the number of clinically significant revealed that 64% of the centers had developed formal policies
adverse events (SAEs) per patient on treatment before and after related to the delivery of cancer care and conduct of clinical tri-
the institution of COVID-19–related policies. In January and als during the COVID-19 pandemic. The policies at many of the
February of 2020, there was a median of 51 SAEs among a me- sites included provisions such as telehealth visits for patients
dian of 2424 patients on treatment (range ¼ 1.76-2.42  10–2 (87.5%), remote work by research staff (75%), and remote moni-
SAEs per patient on treatment). From March to May of 2020, toring by sponsors and/or contract research organizations
there was a median of 30 SAEs among a median of 2347 patients (71.9%), which are also part of the COVID-19 policies at DFCI and
on treatment (range ¼ 1.07-1.92  10–2 SAEs per patient on treat- BH. Even with these modifications, 59.4% of the surveyed cen-
ment). This number decreased to 21 SAEs among 2391 patients ters had halted screening and/or enrollment for clinical trials
on treatment in June 2020. and had ceased research-only visits, except for those providing
Finally, we reviewed patient complaints and did not observe cancer treatments. In addition, roughly one-half of the respond-
any complaints specific to the execution of clinical trials under ing centers had ceased research-only blood draws and/or tissue
the COVID-19–related restrictions. collections, similar to our institutions (16).
The continued enrollment in therapeutic clinical trials at
DFCI and other cancer centers around the country during
COVID-19 demonstrates that cancer clinical research is robust
Discussion
and can be adapted to changing circumstances. Additionally,
The ongoing COVID-19 pandemic is placing tremendous strain these changes can be adopted without compromising safety, as
on the health-care system. Medical and pediatric oncology pro- seen by a lack of increase in serious adverse events and major
grams have been particularly affected due to the ongoing need violations. However, it is prudent to consider the changes taking
to provide optimal care to cancer patients while limiting their place not only as stopgap measures to respond to a temporary
potential for contracting COVID-19 (14,15). In addition to routine health crisis, but perhaps also as considerations in long-term
care, therapeutic clinical trials have also been affected by poli- improvements in clinical trial design to reduce the burden on
cies designed to protect patients and staff from COVID-19. study participants and research staff (17). Many of the modifica-
Despite the limitations imposed by social distancing require- tions made to ensure the safety of patients and staff during
ments, travel restrictions, remote work policies, and limited COVID-19, including telehealth visits, remote monitoring, and
support staff for correlative studies, DFCI was able to continue shipping of investigational agents, could benefit cancer patients
enrolling patients in therapeutic clinical trials at approximately in general as the pandemic wanes. Frequent traveling to the
one-half the pre–COVID-19 rate; moreover, nearly all the clinic in particular has been identified as a source of difficulty
patients previously enrolled were able to stay on the studies. for cancer patients and a barrier to participation in clinical trials
This continued new enrollment and treatment for existing par- (18,19), particularly for ethnic and racial minorities (20,21).
ticipants was mostly due to practice modifications that allowed Reducing the number of required visits for clinical trials would
patients to participate remotely as recommended by our state eliminate unnecessary stress and allow cancer patients to
and federal agencies. These include performing adverse event spend more time with their families, potentially increase the
assessments via telehealth visits and mailing investigational ethnic diversity of clinical trial participants, and improve overall
agents to patients’ homes. Also, more than one-half of the re- quality of life.
search nurses and the majority of CRCs began working remotely COVID-19 is likely to be relevant throughout 2020 and into
once the COVID-19 policies were implemented. This allowed 2021. Even as the threat eases, we should use this situation as
DFCI to maintain a full research staff, albeit under modified an opportunity to reevaluate standard practices for clinical tri-
working conditions. As businesses and other organizations in als. Making clinical trials more accessible, and less stressful for
Massachusetts began to reopen in late May and throughout potential patients, would allow for broader participation. This
June, DFCI began a staged recovery process to resume would have immediate and future benefits for cancer patients
 S. M. Tolaney et al. | 1459

and could streamline the development of effective cancer Mailand, Brenda Acevedo, Sarah Hunt, Sarah Cronholm, Lianne
therapies. Greenspan, Catherine Clinton, Kristen Lawler, Suzanne Ezrre,
Julie Field, Sarah Kelley, Rebecca Rivenburgh, Daniel Surette,
and Loan Vuong. Additionally, the authors would like to ac-
Funding knowledge Timothy K. Erick, PhD, for writing and editorial as-
Steven and Alice Cutler Fund for Clinical Research (to BJ). sistance and Kaitlyn T. Bifolck, BA, for editorial and submission
assistance in the preparation of this manuscript. Both are full-
time employees of Dana-Farber Cancer Institute.
Notes

Role of the funder: The funder had no role in the conception, de- Data Availability
sign, or implementation of this project, analysis of the data, or The datasets generated during and/or analysed during the cur-
preparation of the manuscript. rent study are available from the corresponding author on rea-
sonable request.
Disclosure: SMT receives institutional research funding from
Novartis, Genentech, Eli Lilly, Pfizer, Merck, Exelixis, Eisai,
Bristol Meyers Squibb, AstraZeneca, Cyclacel, References
Immunomedics, Odonate, Sanofi, and Nektar. SMT has 1. Zhu N, Zhang D, Wang W, et al. A novel coronavirus from patients with pneu-
served as an advisor/consultant to Novartis, Eli Lilly, Pfizer, monia in China, 2019. N Engl J Med. 2020;382(8):727-733.

COMMENTARY
2. Coronaviridae Study Group of the International Committee on Taxonomy of V.
Merck, AstraZeneca, Eisai, Puma, Genentech, The species severe acute respiratory syndrome-related coronavirus: classifying
Immunomedics, Nektar, Paxman, Athenex, Oncopep, 2019-nCoV and naming it SARS-CoV-2. Nat Microbiol. 2020;5(4):536-544.
Daiichi-Sankyo, G1 Therapeutics, Silverback Therapeutics, 3. Zhou P, Yang XL, Wang XG, et al. A pneumonia outbreak associated with a
new coronavirus of probable bat origin. Nature. 2020;579(7798):270-273.
Kyowa Kirin Pharmaceuticals, AbbVie, Sanofi, Seattle 4. Holshue ML, DeBolt C, Lindquist S, et al. First case of 2019 novel coronavirus
Genetics, Celldex, Bristol-Myers Squibb and Nanostring. in the United States. N Engl J Med. 2020;382(10):929-936.
SGD received fees for consulting and advisory board roles 5. Johns Hopkins University of Medicine Coronavirus Resource Center. COVID-19
dashboard by the Center for Systems Science and Engineering (CSSE) at Johns
from Bayer and Loxo Oncology and has received travel
Hopkins University (JHU). https://coronavirus.jhu.edu/map.html. Accessed
expenses from Loxo Oncology, Roche/Genentech, and August 20, 2020.
Salarius Pharmaceuticals. JAM reports consulting fees from 6. USAGov. Government Response to Coronavirus, COVID-19. https://www.usa.
gov/coronavirus. Accessed May 11, 2020.
COTA Healthcare, Ignyta, Taiho Pharmaceutical (through
7. Schrag D, Hershman DL, Basch E. Oncology practice during the COVID-19
NCCN grant review panel). MET received honoraria from pandemic. JAMA. 2020;323(20):2005.
Janssen-Ortho, Astellas Pharma, Progenics, AstraZeneca, 8. Marron JM, Joffe S, Jagsi R, et al. Ethics and resource scarcity: ASCO recom-
Clovis Oncology, UpToDate, Guidepoint Global, Merck, mendations for the oncology community during the COVID-19 pandemic. J
Clin Oncol. 2020;38(19):2201-2205.
Bayer, Celegen, Myovant, GSK, Research to Practice, Pfizer, 9. National Cancer Institute. NCI statement on clinical trials during COVID-19
Roivant Sciences, Incyte, and Amgen; consulting/advisory pandemic. https://www.cancer.gov/news-events/press-releases/2020/nci-
role for Guidepoint Global, Research to Practice, Myovant statement-clinical-trials-during-covid-19. Accessed May 11, 2020.
10. U.S. Food and Drug Administration. FDA Guidance on Conduct of Clinical
Sciences, Best Doctors, Inc., Incyte, Arcus (immediate family Trials of Medical Products During COVID-19 Public Health Emergency.
member), Pfizer, Janssen-Ortho, Bayer, UpToDate, Clovis https://www.fda.gov/regulatory-information/search-fda-guidance-docu-
Oncology, Roivant Pharma, AbbieVie, and AstraZeneca; re- ments/fda-guidance-conduct-clinical-trials-medical-products-during-covid-
19-public-health-emergency. Accessed May 11, 2020.
search funding (paid to institution) from Janssen-Ortho, 11. Mass.gov. COVID-19 State of Emergency. https://www.mass.gov/info-details/
Bayer, Pfizer, and Medivation; travel, accommodations, covid-19-state-of-emergency. Accessed May 11, 2020.
expenses from Pfizer, Bayer, Medivation, Incyte, Janssen 12. Brigham and Women’s Hospital. COVID-19 Information and Resources.
https://www.brighamandwomens.org/covid-19. Accessed May 11, 2020.
Oncology, Astellas Pharma, Tokai Pharmaceuticals, and
13. Dana-Farber Cancer Institute. COVID-19 Facts and Resources. https://
Clovis Oncology. All other authors report no disclosures. www.dana-farber.org/covid-19-facts-and-resources/. Accessed May 11,
2020.
Author contributions: Conceptualization: B.E. Johnson, S. 14. Spicer J, Chamberlain C, Papa S. Provision of cancer care during the COVID-19
Tolaney, M.E. Taplin, J.A. Meyerhardt, Data curation: C.A. Lydon, pandemic. Nat Rev Clin Oncol. 2020;17(6):329-331.
15. Liang W, Guan W, Chen R, et al. Cancer patients in SARS-CoV-2 infection: a
J. Dai, A. Strandring, K. Legor, C.M. Caparrotta, M. Schenker, D.I.
nationwide analysis in China. Lancet Oncol. 2020;21(3):335-337.
Glazer, Formal analysis: N. Tayob, T. Li, C.A. Lydon, Funding ac- 16. Waterhouse DM, Harvey RD, Hurley P, et al. Early impact of COVID-19 on the
quisition: B.E. Johnson, Investigation: B.E. Johnson, S. Tolaney, conduct of oncology clinical trials and long-term opportunities for transfor-
mation: findings from an American Society of Clinical Oncology Survey. J Clin
M.E. Taplin, J.A. Meyerhardt, Methodology: B.E. Johnson, S.
Oncol Pract. 2020;16(7):417–421.
Tolaney, M.E. Taplin, J.A. Meyerhardt, Project administration: 17. Nabhan C, Choueiri TK, Mato AR. Rethinking clinical trials reform during the
C.A. Lydon, Resources: B.E. Johnson, Writing—original draft: B.E. COVID-19 pandemic. JAMA Oncol. 2020; 6(9):1327–1329.
Johnson, S. Tolaney Writing—review and editing: all authors. 18. Avis NE, Smith KW, Link CL, et al. Factors associated with participation in
breast cancer treatment clinical trials. J Clin Oncol. 2006;24(12):1860-1867.
19. Nipp RD, Lee H, Powell E, et al. Financial burden of cancer clinical trial partici-
Acknowledgments: The authors wish to acknowledge the re-
pation and the impact of a cancer care equity program. Oncologist. 2016;21(4):
search managers who replied to the survey: Nicole Ryabin, 467-474.
Leilani Anderson, Janice Russell, Sarah Garcia, Ketki Bhushan, 20. Hamel LM, Penner LA, Albrecht TL, et al. Barriers to clinical trial enrollment
Michele Vincitore, Brittney Wilkinson, Meredith Flynn, Kim in racial and ethnic minority patients with cancer. Cancer Control. 2016;23(4):
327-337.
MacNeill, Julie Childress, Farzana Masood, Frank Brown, Kelsey 21. Nipp RD, Hong K, Paskett ED. Overcoming barriers to clinical trial enrollment.
O’Day, Jillian Foreman, Jen Maattala, Diane Warren, Deb Am Soc Clin Oncol Educ Book. 2019;39(39):105-114.