Key Points
Question
Does a shared model of care after allogeneic hematopoietic cell transplantation (HCT) between HCT specialists and local oncologists preserve nonrelapse mortality and improve quality of life?
Findings
This randomized clinical trial including 302 patients who underwent HCT found that a model of post-HCT care delivery, engaging and training local oncologists to share patient care with HCT specialists, did not compromise day 100 nonrelapse mortality and led to improved quality of life at day 100 but not day 180.
Meaning
These results indicate that shared care after HCT is likely safe and able to reduce early patient and caregiver burdens.
Abstract
Importance
Although sharing care with local oncologists after allogeneic hematopoietic cell transplantation (HCT) has been proposed for patients living far from HCT centers, it is not known whether a shared strategy is safe or improves patient quality of life (QOL).
Objective
To determine the efficacy and safety of sharing follow-up care after HCT between the HCT specialty center and local oncologists.
Design, Setting, and Participants
This was a multicenter collaborative randomized clinical trial of patients undergoing HCT at Dana-Farber Cancer Institute (DFCI)—a high volume HCT center in Boston (Massachusetts)—and 8 local oncology practices. Eligible patients were enrolled from December 2017 to December 2021 and were randomized 1:1 to shared vs usual care after neutrophil engraftment, stratified by local sites in Massachusetts, Rhode Island, New Hampshire, New York, and Maine. Data analyses were performed in January 2024.
Intervention
Shared care involved alternating post-HCT visits at DFCI and local oncology practices through day 100; for usual care, all post-HCT visits occurred only at DFCI.
Main Outcomes and Measures
Coprimary outcomes were nonrelapse mortality (NRM) at day 100, and QOL measured by the FACT-BMT (Functional Assessment of Cancer Therapy–Bone Marrow Transplantation) instrument and the QLQ-C30 (European Organization for Research and Treatment of Cancer’s Quality of Life Questionnaire) at day 180. Prespecified secondary outcomes included day 100 QOL and 1-year overall survival.
Results
A total of 302 participants (median [range] age, 63 [20-79] years; 117 [38.7%] females; 185 [61.3%] males) were included in the analysis; 152 were randomized to shared care and 150 to usual care. Day 100 NRM was noninferior for shared vs usual care (2.6% [95% CI, 0.7% to 6.6%] vs 2.7% [95% CI, 0.7% to 6.7%]; P = .98). There were no differences at day 180 for the FACT-BMT total score (mean difference, 3.8; 95% CI, −2.1 to 9.6; P = .20) or QLQ-C30 global score (1.9; 95% CI, −4.9 to 8.8; P = .58). At day 100, the FACT-BMT total score was better for shared care (mean difference, 6.6; 95% CI, 1.0 to 12.1; P = .02) as was the QLQ-C30 global score (8.8; 95% CI, 1.8 to 15.7; P = .02).
Conclusions and Relevance
This randomized clinical trial found that shared care resulted in noninferior NRM at day 100 but similar QOL at day 180, with improved QOL at day 100. These data suggest that shared care is safe, improves QOL early on, and has the potential to become a routine model for post-HCT care.
Trial Registration
ClinicalTrials.gov Identifier: NCT03244826
This randomized clinical trial evaluates the efficacy of sharing follow-up care between the specialty center and a local oncology practice after allogeneic hematopoietic cell transplantation.
Introduction
Allogeneic hematopoietic cell transplantation (HCT) is a potentially curative treatment for many hematologic malignant neoplasms.1,2,3 Methodologic advances in human leukocyte antigen typing, graft vs host disease (GVHD) prevention strategies, and supportive care have increased the number of patients undergoing HCT; however, it remains a highly technical procedure available only at select centers that can collect, process, and store hematopoietic stem cells, as well as care for patients before immune function recovers. Moreover, although many eventually return to baseline health, patients who undergo HCT are at particular risk of short- and long-term adverse effects of the procedure itself. These include infections (both rare and common) and acute and chronic GVHD, which in addition to concerns for disease relapse, require diligent follow-up by the HCT team after discharge from the initial inpatient procedure.
Usual care for patients after HCT at most centers consists of weekly visits for several months to monitor for infections and GVHD; tailor immune suppression drug levels; and provide potential transfusions. This is difficult for patients and their caregivers who live at great distances from the HCT center, adding emotional and/or financial stress that often impairs quality of life (QOL) and financial well-being.4,5,6 On the other hand, it is not known whether complex early post-HCT care can safely be shared between HCT specialists and local oncologists without compromising HCT outcomes.
We assessed the safety and efficacy of a post-HCT shared care program. This approach allowed a randomized cohort of patients to receive approximately half of their early post-HCT care at the HCT center, and the other half with their local oncologist. We hypothesized that sharing care would lead to similar day 100 (D100) nonrelapse mortality (NRM) as well as improvement in QOL at day 180 (D180).
Methods
This study was reviewed and approved by the Dana-Farber−Harvard Cancer Center Office for Human Subjects Research and by each local site’s institutional review board. The trial protocol is available in Supplement 1. Patients were informed about the study when they first presented to DFCI for transplant consultation, and if interested, written informed consent was obtained on the day of clinical HCT consent. The study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline.
Study Setting
We undertook a multisite randomized clinical trial at Dana-Farber Cancer Institute (DFCI) in Boston (Massachusetts) and across 8 local oncology practices throughout the New England region and upstate New York. These included approximate driving times to Boston from Dartmouth-Hitchcock Medical Center in Lebanon, New Hampshire (2.5 hours); New York Oncology Hematology in Albany, New York (3 hours); New England Cancer Specialists in Brunswick, Maine (2 hours); Northern Light Cancer Care in Brewer, Maine (4 hours); Lifespan Cancer Institute in Providence, Rhode Island (1.5 hours); Dana-Farber South Shore in Weymouth, Massachusetts (40 min); Dana-Farber Milford in Milford, Massachusetts (1 hour); and Dana-Farber Londonderry in Londonderry, New Hampshire (1.5 hours). Patients planning for allogeneic HCT from December 2017 to December 2021 were enrolled and randomized to a study arm before inpatient admission.
Participants
We included patients planning to undergo allogeneic HCT at DFCI who were aged 18 years or older; able to speak and read English; living in Connecticut, Massachusetts, Maine, New Hampshire, New York, or Vermont; living 30 minutes or farther from Boston; and able to provide informed consent. Demographic information available in the medical record was used. Exclusions included participation in another clinical trial requiring all post-HCT monitoring at DFCI, cord blood HCT, or previous allogeneic transplantation.
Intervention
Patients signed informed consent on the day of HCT consent (approximately 1 week before admission for HCT), and shortly after were randomized 1:1 to each study arm (Figure 1). During the first 100 days after HCT, patients in the shared care arm received approximately half of the follow-up visits at a local oncology practice, alternating between a local oncologist and the HCT team at DFCI. The model involved 4 strategies for safe local care: (1) an online care coordination plan, (2) patient engagement and education, (3) local oncologist engagement and education, and (4) a web-based communications platform that patients, local oncologists, and transplant specialists all had access to (eFigure 1 in Supplement 2). Patients in the usual care group received the standard protocol: all follow-up visits until D100 were with the HCT team at DFCI in Boston. Randomization was stratified by local center, and shared care began with the first outpatient clinic visit after neutrophil engraftment.
Figure 1. CONSORT Diagram of Study Participants.
D100 indicates day 100; D180, day 180; FACT-BMT, Functional Assessment of Cancer Therapy–Bone Marrow Transplantation; HCT, hematopoietic cell transplantation; NRM, nonrelapse mortality; PROs, patient-reported outcomes; and QLQ-30, Quality of Life Questionnaire.
aNonrandomized participants were not enrolled but agreed to be followed up for PROs forming a reference cohort to assess whether patients who agreed to be in the study were different from patients who did not. They received standard of care (all post-HCT in Boston). More details are available in the eAppendix and eTable 5 in Supplement 2.
Two shared-care oncologists were present at each local site; training for these physicians occurred at DFCI before the study initiation, and then annually in Boston (eAppendix in Supplement 2), including topics such as optimal infection monitoring, new HCT techniques, and grading and management of GVHD. Training was also provided for local advanced practice practitioners, and the availability of local specialized HCT laboratory capabilities was ensured. A dedicated DFCI shared-care attending physician was available via pager at all hours for consultation.
Outcomes and Data Collection
Coprimary outcomes were NRM at D100 and QOL at D180 as assessed by the Functional Assessment of Cancer Therapy–Bone Marrow Transplantation (FACT-BMT)7 instrument and the European Organization for Research and Treatment of Cancer’s Quality of Life Questionnaire (QLQ-C30).8 Prespecified secondary outcomes included acute GVHD, QOL at D100, and 1-year overall survival (OS). In addition to standard clinical data collection such as NRM and OS (required by federal law for all patients who undergo HCT in the US), patients were asked to complete patient-reported outcomes (PROs) at D100 and D180 (with instruments chosen from a modified Delphi panel of patients with HCT before the study).9 The selected PROs (FACT-BMT and QLQ-C308,10) focused on physical, social, emotional, and financial11 well-being (eAppendix in Supplement 2).
Statistical Analysis
Data analysis was based on the intent-to-treat principle for HCT outcome measures. T tests were used to assess differences in QOL measures at D180 and D100, and multivariable analyses were performed for FACT-BMT at D180 and D100. Time-to-event analyses were performed for OS and NRM at D100. For OS and progression-free survival, standard survival analyses were performed, including Kaplan-Meier methods for estimation of survival functions and log-rank tests for group comparison of survival curves. For NRM, relapse, and acute GVHD, competing risks data analyses were performed. Cumulative incidences were compared using the Gray test.12 Because these end points are immunologically intertwined, they were analyzed in a competing risks framework.13,14
All P values were 2-sided, and P ≤ .05 was considered statistically significant; multiplicity was not considered. Data analyses were performed using SAS, version 9.4 (SAS Institute Inc) and R, version 3.5.1 (the R Foundation for Statistical Computing) in January 2024.
Results
Cohort Characteristics
A total of 302 patients (median [range] age, 63 [20-79] years; 117 females [38.7%] and 185 males [61.3%]) were enrolled and received HCT, with 152 randomized to shared care and 150 to usual care (Table 1). The cohort included 2 American Indian/Alaska Native (0.7%), 6 Asian (2%), 5 Black (1.7%), 277 White (91.7%), and 12 individuals of another race or unknown (4%). There were no significant differences between baseline sociodemographic and clinical factors in the cohorts except for hematopoietic progenitor cell source. For shared care, the median (range) number of local visits was 5 (1-17), and DFCI visits, 6 (1-19); for usual care these were 0 (0-3) and 11 (6-18) visits, respectively.
Table 1. Characteristics of Patients Undergoing HCT Followed by Shared or Usual Care.
| Characteristic | No. (%) | P value | ||
|---|---|---|---|---|
| All | Shared care | Usual care | ||
| Total participants, No. | 302 | 152 (50.3) | 150 (49.7) | NA |
| Age at HCT, median (range), y | 63 (20-79) | 63 (20-79) | 62 (20-77) | .64 |
| Female | 117 (38.7) | 53 (34.9) | 64 (42.7) | .19 |
| Male | 185 (61.3) | 99 (65.1) | 86 (57.3) | |
| Patient-donor sex mismatch | ||||
| Female donor to male patient | 65 (21.5) | 35 (23.0) | 30 (20.0) | .58 |
| Karnofsky score, points | ||||
| 100-90 | 103 (34.1) | 49 (32.2) | 54 (36.0) | .54 |
| 70-80 | 199 (65.9) | 103 (67.8) | 96 (64.0) | |
| Race | ||||
| American Indian/Alaska Native | 2 (0.7) | 1 (0.7) | 1 (0.7) | .42 |
| Asian | 6 (2.0) | 2 (1.3) | 4 (2.7) | |
| Black | 5 (1.7) | 4 (2.6) | 1 (0.7) | |
| White | 277 (91.7) | 141 (92.8) | 136 (90.7) | |
| Another race or unknowna | 12 (4.0) | 4 (2.6) | 8 (5.3) | |
| Disease | ||||
| ALL | 41 (13.6) | 20 (13.2) | 21 (14.0) | .86 |
| AML | 126 (41.7) | 66 (43.4) | 60 (40.0) | |
| Atypical CML | 1 (0.3) | 0 | 1 (0.7) | |
| CML | 3 (1.0) | 2 (1.3) | 1 (0.7) | |
| human leukocyte | 2 (0.7) | 0 | 2 (1.3) | |
| Hemoglobinopathy | 2 (0.7) | 1 (0.7) | 1 (0.7) | |
| MDS | 71 (23.5) | 36 (23.7) | 35 (23.3) | |
| MDS/MPN | 1 (0.3) | 0 | 1 (0.7) | |
| MPN | 22 (7.3) | 12 (7.9) | 10 (6.7) | |
| NHL | 20 (6.6) | 9 (5.9) | 11 (7.3) | |
| Other acute leukemia | 3 (1.0) | 1 (0.7) | 2 (1.3) | |
| Other leukemia | 6 (2.0) | 4 (2.6) | 2 (1.3) | |
| SAA | 4 (1.3) | 1 (0.7) | 3 (2.0) | |
| HCT comorbidity index | ||||
| 0-2 | 139 (46.0) | 63 (41.0) | 76 (51.0) | .10 |
| 3-6 | 143 (47.0) | 81 (53.0) | 62 (41.0) | |
| ≥7 | 20 (7.0) | 8 (5.0) | 12 (8.0) | |
| Patient-donor CMV sero status, positive | 168 (55.6) | 78 (51.3) | 90 (60.0) | .13 |
| Cell source | ||||
| PBSC | 270 (89.4) | 142 (93.4) | 128 (85.3) | .03 |
| Bone marrow | 32 (10.6) | 10 (6.6) | 22 (14.7) | |
| Conditioning intensityb | .80 | |||
| MAC | 94 (31.1) | 46 (30.3) | 48 (32.0) | NAc |
| Cy/TBI | 14 | 6 | 8 | |
| Flu/Bu4 | 67 | 32 | 35 | |
| Flu/Mel | 1 | 1 | 0 | |
| Flu/Mel/AT | 1 | 1 | 0 | |
| Flu/TBI | 4 | 2 | 2 | |
| TBI/Cy | 7 | 4 | 3 | |
| RIC | 208 (68.9) | 106 (69.7) | 102 (68) | NAc |
| Flu/Bu1 | 12 | 4 | 8 | |
| Flu/Bu2 | 136 | 70 | 66 | |
| Flu/Cy | 3 | 1 | 2 | |
| Flu/Cy/TBI | 2 | 0 | 2 | |
| Flu/Cy/TLI | 2 | 2 | 0 | |
| Flu/Mel | 30 | 16 | 14 | |
| Flu/TBI/Cy | 23 | 13 | 10 | |
| HLA type (A, B, C, DRB1) | ||||
| Matched unrelated | 182 (60.3) | 87 (57.2) | 95 (63.3) | .37 |
| Matched related | 45 (14.9) | 27 (17.8) | 18 (12.0) | |
| 7/8 Matched unrelated | 38 (12.6) | 20 (13.2) | 18 (12.0) | |
| 7/8 Matched related | 2 (0.7) | 0 | 2 (1.3) | |
| Haploidentical | 35 (11.6) | 18 (11.8) | 17 (11.3) | |
| GVHD prophylaxis | ||||
| Tac/MTX | 146 (48.3) | 72 (47.4) | 74 (49.3) | .36 |
| Tac/MTX/Rap | 60 (19.9) | 35 (23.0) | 25 (16.7) | |
| Tac/Rap | 15 (5.0) | 7 (4.6) | 8 (5.3) | |
| PTCY-based | 79 (26.2) | 36 (23.7) | 43 (28.7) | |
| Other | 2 (0.7) | 2 (1.3) | 0 | |
Abbreviations: ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; Bu, busulfan; CML, chronic myeloid leukemia; CMV, cytomegalovirus; Cy, cyclophosphamide; Flu, fludarabine; GVHD, graft vs host disease; HCT, hematopoietic cell transplantation; HLA, human leukocyte antigen; NA, not applicable; MAC, myeloablative conditioning; MDS, myelodysplastic syndromes; Mel, melphalan; MPN, myeloproliferative neoplasm; MTX, methotrexate; NHL, non-Hodgkin lymphoma; PBSC, peripheral blood stem cells; PCTY, posttransplant cyclophosphamide; Rap, sirolimus; RIC, reduced-intensity conditioning; SAA, severe aplastic anemia; Tac, tacrolimus; TBI, total body irradiation.
Race was collected from the electronic medical records and was self-reported.
Conditioning intensity was defined by standard criteria. RIC regimens included Flu, 120 mg/m2, with intravenous Bu at doses of 3.2 mg/kg (Flu/Bul) or 6.4 mg/kg (Flu/Bu2); Flu, 120 mg/m2, with Mel, 100-140 mg/m2 (Flu/Mel); or Flu with Cy and low-dose TBI, 200 cGy (Flu/Cy/TBI). MAC regimens included high-dose Bu, 12.8 mg/kg, with Flu (Flu/Bu4); Cy, 3600 mg/m2, with TBI, 1200 cGy (Cy/TBI); Flu, 90 mg/m2, with TBI, 1200 cGy (Flu/TBI); and Flu, 120 mg/m2, with Mel, 140 mg/m2, and Bu, 9.6 mg/kg (Flu/Bu/Mel).
Too many categories with small numbers so estimates do not converge and were intentionally excluded.
Primary Outcomes
On D100, NRM was 2.6% for shared care and 2.7% for usual care. The upper limit of the 1-sided 90% CIs for this difference was 2.3%, below the prespecified noninferiority margin of 5.0%. At D100, 286 randomized participants (94.7%) were alive. There were no significant differences in OS (94% [95% CI, 89%-97%] vs 95% [95% CI, 90%-98%]; P = .62) or NRM (2.7% [95% CI, 0.9%-6.0%] vs 2.7% [95% CI, 0.9%-6.0%]; P = .98) for shared vs usual care (Figure 2).
Figure 2. Nonrelapse Mortality (NRM) and Overall Survival (OS) at Day 100 With Shared vs Usual Care After Hematopoietic Cell Transplantation.
At D180, 260 participants (85.5%) were alive, with complete FACT-BMT data available on 182 participants (94 shared and 88 usual care; Table 2), for a response rate of 70.0%. Response rates were similar between groups and across subscales. Shared-care participants had a significantly higher mean score on the physical well-being subscale significantly higher mean score on the physical well-being subscale (1.9; 95% CI, 0.3-3.5; P = .02) and higher but nonsignificant scores on other subscales (Table 2; multivariable analysis is available in eFigure 2 and eTable 1 in Supplement 2). There were no significant differences observed on the QLQ-C30 global score or subscales.
Table 2. Functional Assessment of Cancer Therapy–Bone Marrow Transplantation (FACT-BMT) Instrument at Day 180 and at Day 100.
| Day and FACT-BMT scale | Shared care (n = 94) | Usual care (n = 88) | P valuea | ||
|---|---|---|---|---|---|
| Patients, No. | Mean (SD) score | Patients, No. | Mean (SD) score | ||
| Day 180 | |||||
| PWB | 92 | 22.5 (4.7) | 88 | 20.6 (6.3) | .02 |
| SWB | 91 | 23.9 (4.0) | 88 | 24 (3.5) | .82 |
| EWB | 91 | 19.2 (3.8) | 87 | 18.8 (4.1) | .54 |
| FWB | 89 | 17.0 (5.8) | 87 | 16.7 (6.2) | .71 |
| FACT-G | 88 | 82.7 (13.7) | 86 | 80.4 (14.6) | .29 |
| BMTS | 89 | 28.2 (5.7) | 87 | 26.4 (6.9) | .06 |
| TOI | 86 | 67.7 (14.0) | 87 | 63.8 (16.8) | .10 |
| TOTAL | 86 | 110.7 (18.3) | 86 | 106.9 (20.4) | .20 |
| Day 100 | |||||
| PWB | 90 | 22.1 (4.5) | 83 | 19.6 (6.2) | .004 |
| SWB | 89 | 23.4 (4.1) | 82 | 23.2 (4.0) | .73 |
| EWB | 88 | 19.6 (3.7) | 83 | 18.4 (3.9) | .04 |
| FWB | 87 | 15.5 (5.8) | 83 | 14.9 (5.7) | .50 |
| FACT-G | 87 | 80.5 (13.0) | 82 | 76.3 (15.0) | .06 |
| BMTS | 88 | 28.4 (4.9) | 82 | 26.22 (5.6) | .007 |
| TOI | 87 | 66.0 (12.8) | 82 | 60.5 (15.0) | .01 |
| TOTAL | 87 | 108.9 (16.8) | 81 | 102.4 (19.4) | .02 |
Abbreviations: BMTS, Bone Marrow Transplant subscale; EWB, emotional well-being; FACT-G, Functional Assessment of Cancer Therapy-General; FWB, functional well-being; PWB, physical well-being; SWB, social well-being; TOI, FACT-BMT Trial Outcome Index; TOTAL, FACT-General plus BMT.
Calculated using t tests.
Secondary Outcomes, Financial Hardship, and Resource Utilization
At D100, 286 participants (94.4%) were alive; of these, complete FACT-BMT data were available for 173 participants (response rate of 60.5%). Shared-care participants reported higher mean scores on all 5 subscales (physical, social/family, emotional, and functional well-being, and the bone marrow transplantation subscale (BMTS) and all 3 composite scores (the FACT-General, Trial Outcome Index, and the FACT-General plus BMT total; Table 2; multivariable analysis in eFigure 2 and eTable 2 in Supplement 2). The difference in BMTS (2.2 points higher for shared care) exceeded the lower bound of the minimally clinically important difference (MCID) of 2 to 3 points as described by its developers.7
On the QLQ-C30 at D100, shared-care patients reported higher mean functional scale scores and lower mean symptom scale scores. The difference in the QLQ-C30 global score for shared care (8.8 points higher; 95% CI, 1.8-15.7; P = .02) exceeded the MCID of 6.9 points higher for global QOL among patients with HCT on the QLQ-C30 as described by Hong et al.15
At D100, cumulative incidence of all-grade acute GVHD was 30.9% (95% CI, 23.7%-38.9%) for shared care and 31.3% (95% CI, 24.0% vs 39.4%) for usual care (P = .99). Cumulative incidence of grade II-IV acute GVHD was also similar (17.1% [95% CI, 11.5%-24.1%] vs 16.0% [95% CI, 10.5%-22.9%]; P = .72).
At 1 year, OS was 71% (95% CI, 63%-78%) vs 75% (95% CI, 67%-81%) for shared care and usual care, respectively (P = .48), progression-free survival was 61% (95% CI, 53%-68%) vs 64% (95% CI, 56%-72%; P = .51), cumulative incidence of NRM was 10.6% (95% CI, 6.3%-16%) vs 9.4% (95% CI, 5.4%-15%; P = .70), and cumulative incidence of relapse was 29% (95% CI, 22%-36%) vs 26% (95% CI, 19%-33%; P = .67; eFigure 3 in Supplement 2). In multivariable analysis, the hazard ratio for death within 1 year for shared vs usual care was 1.11 (95% CI, 0.71-1.74; eTable 3 in Supplement 2).
No significant differences were observed in measures of financial hardship at D100 or D180 for shared vs usual care (additional analyses are available in the eAppendix in Supplement 2). Inpatient readmissions were similar for shared vs usual care (eTable 4 in Supplement 2).
Discussion
Although offering a potential cure, for patients who live far from the HCT specialty center, allogeneic HCT creates profound financial, psychosocial, and physical burdens (ie, relocation or repeated travel) that impact patients’ health beyond the hematologic disorder. This randomized clinical study confirmed part of our hypothesis: a model of post-HCT care shared with local oncologists was not associated with inferior NRM at D100. In contrast, we found that QOL was similar at D180 for shared vs usual care. Interestingly, the shared model was associated with improvement in QOL sooner in the post-HCT period (D100), with consistent enhancements across 2 validated measures and multiple parallel domains.
Shared care models have been used in the oncology setting mostly for primary care practitioners to partner with oncologists. This includes patients newly diagnosed,16 undergoing chemotherapy,17,18 or who are cancer survivors returning to the community (prostate,19 childhood,20 and colorectal cancer21). Our model, in which subspecialty oncology care was shared with general oncologists, is more rare and has lacked rigorous evaluation.22 An exception is a Canadian program for acute myeloid leukemia that retrospectively analyzed 73 patients receiving shared postconsolidation care at local hospitals vs 344 patients receiving care only at the quaternary center.23 Patients in the shared care arm saved a mean travel distance of 146.5 km per local visit, with no differences in OS. Their burden of care was undoubtedly lightened.
Indeed, a study of allogeneic HCT survivors found that 47% reported ongoing financial burden after HCT.24 Moreover, in a multisite survey of adult patients 180 days after HCT (n = 325), income decline was reported by 46% of patients; 57% reported financial hardship after the procedure; and 16% reported extreme hardship.4 In turn, financial hardship was associated with reporting QOL below the median and perceived stress above the median. These data suggest that among the reasons that QOL was better at D100 for shared-care patients in the current study could have been through improved finances, although we did not find a difference in financial hardship at that time point. This may be because our hardship questions were not sensitive enough, or because shared care provided nonfinancial psychosocial benefits that we did not capture.
Shared care required substantial and ongoing coordination between specialty HCT centers and local oncologists, and this approach must be taken carefully to protect patients from early release to local clinicians with inadequate resources to care for them. Extensive initial training was also necessary to familiarize the local oncologists with the needs of patients with HCT. These may limit the strategy’s applicability in resource-limited settings. In addition, although we started the study with a web-based communication platform, it was somewhat onerous because it was not integrated with the electronic health record (EHR). Over the years of the study, the web-based platform was mostly replaced by the “care everywhere” function in DFCI’s Epic EHR, which allows entry into other linked EHRs throughout the region.
Limitations
There were limitations to our analysis. The study was not designed to detect changes in QOL at D100, and survey response rates were lower at this time point. Additionally, given shared care is a care delivery strategy, it was not possible to mandate that patients receive shared care every other week. Moreover, the intention-to-treat analytical approach, a conservative strategy with respect to treatment efficacy, may have underestimated the association of shared care with QOL. We did have a small difference in stem cell sources between the 2 groups, and also, the study cohort was composed of mostly adults and White individuals, which could affect generalizability. Next, although we continued to enroll participants during the COVID-19 pandemic, the survey follow-up and study meetings required a delicate balance between maintaining study rigor while remaining flexible to shift according to public health conditions. Finally, while we included patients from rural areas and with different insurance types, shared care may be difficult to apply in more remote communities or ones with additional locoregional disparities in access to care.
Conclusions
This randomized clinical trial found that shared care with local oncologists has the potential to become a routine model for follow-up care after allogeneic HCT, both in the US and in other parts of the world where HCT centers treat patients who live far enough away that weekly commuting is burdensome. Our data suggest it is safe, improves patients’ lived experiences in the early post-HCT period and has the advantage of potentially allowing overburdened specialized centers to treat more patients by off-loading some volume of post-HCT care to local practices. Although the shared care model may create local burdens that will also need to be addressed, partial decentralization of post-HCT care has the potential to improve care equity for patients who live far from specialized oncology centers.
Trial Protocol
eAppendix.
eFigure 1. Overview of the Shared Care Model
eFigure 2. Unadjusted and Adjusted FACT-BMT (BMTS) Box Plots, D100 and Day 180
eFigure 3. Overall Survival, Progression-Free Survival, Non-Relapse Mortality and Relapse at One Year after HCT
eTable 1. Multivariable FACT-BMT at D180
eTable 2. Multivariable FACT-BMT at D100
eTable 3. Multivariable Analysis of HCT Outcomes at 1 Year
eTable 4. Frequency of Readmissions
eTable 5. Characteristics of Non-Randomized versus Randomized Cohorts
Data Sharing Statement
References
- 1.Ljungman P, Urbano-Ispizua A, Cavazzana-Calvo M, et al. ; European Group for Blood and Marrow . Allogeneic and autologous transplantation for haematological diseases, solid tumours and immune disorders: definitions and current practice in Europe. Bone Marrow Transplant. 2006;37(5):439-449. doi: 10.1038/sj.bmt.1705265 [DOI] [PubMed] [Google Scholar]
- 2.Copelan EA. Hematopoietic stem-cell transplantation. N Engl J Med. 2006;354(17):1813-1826. doi: 10.1056/NEJMra052638 [DOI] [PubMed] [Google Scholar]
- 3.Gooley TA, Chien JW, Pergam SA, et al. Reduced mortality after allogeneic hematopoietic-cell transplantation. N Engl J Med. 2010;363(22):2091-2101. doi: 10.1056/NEJMoa1004383 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Abel GA, Albelda R, Khera N, et al. Financial hardship and patient-reported outcomes after hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2016;22(8):1504-1510. doi: 10.1016/j.bbmt.2016.05.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Beattie S, Lebel S. The experience of caregivers of hematological cancer patients undergoing a hematopoietic stem cell transplant: a comprehensive literature review. Psychooncology. 2011;20(11):1137-1150. doi: 10.1002/pon.1962 [DOI] [PubMed] [Google Scholar]
- 6.Polomeni A, Lapusan S, Bompoint C, Rubio MT, Mohty M. The impact of allogeneic-hematopoietic stem cell transplantation on patients’ and close relatives’ quality of life and relationships. Euro J Oncol Nurs. 2016;21:248-256. 2015;doi: 10.1016/j.ejon.2015.10.011 [DOI] [PubMed] [Google Scholar]
- 7.McQuellon RP, Russell GB, Cella DF, et al. Quality of life measurement in bone marrow transplantation: development of the Functional Assessment of Cancer Therapy-Bone Marrow Transplant (FACT-BMT) scale. Bone Marrow Transplant. 1997;19(4):357-368. doi: 10.1038/sj.bmt.1700672 [DOI] [PubMed] [Google Scholar]
- 8.Aaronson NK, Ahmedzai S, Bergman B, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst. 1993;85(5):365-376. doi: 10.1093/jnci/85.5.365 [DOI] [PubMed] [Google Scholar]
- 9.Solle JC, Steinberg A, Marathe P, Gray TF, Emmert A, Abel GA. Patients as experts: characterizing the most relevant patient-reported outcomes after hematopoietic cell transplantation. Bone Marrow Transplant. 2020;55(1):242-244. doi: 10.1038/s41409-019-0518-0 [DOI] [PubMed] [Google Scholar]
- 10.Grulke N, Albani C, Bailer H. Quality of life in patients before and after haematopoietic stem cell transplantation measured with the European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Core Questionnaire QLQ-C30. Bone Marrow Transplant. 2012;47(4):473-482. doi: 10.1038/bmt.2011.107 [DOI] [PubMed] [Google Scholar]
- 11.Abel G, Salas-Coronado D, Albelda R, et al. Prospective assessment of familial financial hardship after hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2015;21:572. doi: 10.1016/j.bbmt.2014.11.078 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Gray RJ. Class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat. 1988;16(3):1141-1154. doi: 10.1214/aos/1176350951 [DOI] [Google Scholar]
- 13.Kim HT. Cumulative incidence in competing risks data and competing risks regression analysis. Clin Cancer Res. 2007;13(2 Pt 1):559-565. doi: 10.1158/1078-0432.CCR-06-1210 [DOI] [PubMed] [Google Scholar]
- 14.Kim HT, Armand P. Clinical endpoints in allogeneic hematopoietic stem cell transplantation studies: the cost of freedom. Biol Blood Marrow Transplant. 2013;19(6):860-866. doi: 10.1016/j.bbmt.2013.01.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hong F, Bosco JL, Bush N, Berry DL. Patient self-appraisal of change and minimal clinically important difference on the European Organization for the Research and Treatment of Cancer quality of life questionnaire core 30 before and during cancer therapy. BMC Cancer. 2013;13:165. doi: 10.1186/1471-2407-13-165 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Nielsen JD, Palshof T, Mainz J, Jensen AB, Olesen F. Randomised controlled trial of a shared care programme for newly referred cancer patients: bridging the gap between general practice and hospital. Qual Saf Health Care. 2003;12(4):263-272. doi: 10.1136/qhc.12.4.263 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Jefford M, Baravelli C, Dudgeon P, et al. Tailored chemotherapy information faxed to general practitioners improves confidence in managing adverse effects and satisfaction with shared care: results from a randomized controlled trial. J Clin Oncol. 2008;26(14):2272-2277. doi: 10.1200/JCO.2007.14.7710 [DOI] [PubMed] [Google Scholar]
- 18.Johnson CE, Saunders CM, Phillips M, et al. Randomized controlled trial of shared care for patients with cancer involving general practitioners and cancer specialists. J Oncol Pract. 2015;11(5):349-355. doi: 10.1200/JOP.2014.001569 [DOI] [PubMed] [Google Scholar]
- 19.Emery JD, Jefford M, King M, et al. ProCare Trial: a phase II randomized controlled trial of shared care for follow-up of men with prostate cancer. BJU Int. 2017;119(3):381-389. doi: 10.1111/bju.13593 [DOI] [PubMed] [Google Scholar]
- 20.Blaauwbroek R, Tuinier W, Meyboom-de Jong B, Kamps WA, Postma A. Shared care by paediatric oncologists and family doctors for long-term follow-up of adult childhood cancer survivors: a pilot study. Lancet Oncol. 2008;9(3):232-238. doi: 10.1016/S1470-2045(08)70034-2 [DOI] [PubMed] [Google Scholar]
- 21.Jefford M, Emery J, Grunfeld E, et al. SCORE: shared care of colorectal cancer survivors: protocol for a randomised controlled trial. Trials. 2017;18(1):506. doi: 10.1186/s13063-017-2245-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Cheung V, Siddiq N, Devlin R, McNamara C, Gupta V. Shared-care model for complex chronic haematological malignancies. Can Oncol Nurs J. 2021;31(2):165-174. doi: 10.5737/23688076312165174 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hershenfeld SA, Maki K, Rothfels L, et al. Sharing post-AML consolidation supportive therapy with local centers reduces patient travel burden without compromising outcomes. Leuk Res. 2017;59:93-96. doi: 10.1016/j.leukres.2017.05.023 [DOI] [PubMed] [Google Scholar]
- 24.Khera N, Chang YH, Hashmi S, et al. Financial burden in recipients of allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2014;20(9):1375-1381. doi: 10.1016/j.bbmt.2014.05.011 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Trial Protocol
eAppendix.
eFigure 1. Overview of the Shared Care Model
eFigure 2. Unadjusted and Adjusted FACT-BMT (BMTS) Box Plots, D100 and Day 180
eFigure 3. Overall Survival, Progression-Free Survival, Non-Relapse Mortality and Relapse at One Year after HCT
eTable 1. Multivariable FACT-BMT at D180
eTable 2. Multivariable FACT-BMT at D100
eTable 3. Multivariable Analysis of HCT Outcomes at 1 Year
eTable 4. Frequency of Readmissions
eTable 5. Characteristics of Non-Randomized versus Randomized Cohorts
Data Sharing Statement


