This randomized clinical trial assesses whether geriatric assessment–driven intervention can reduce chemotherapy-related toxic effects in older adults with cancer.
Key Points
Question
Can geriatric assessment–driven intervention reduce grade 3 or higher chemotherapy-related toxic effects in older adults with cancer?
Findings
In this randomized clinical trial that included 605 eligible older adults with cancer starting a new chemotherapy regimen, incidence of grade 3 or higher chemotherapy-related toxic effects was 50.5% in patients receiving geriatric assessment–driven intervention vs 60.6% in patients receiving standard of care, resulting in a significant 10.1% reduction.
Meaning
Integration of a multidisciplinary geriatric assessment–driven intervention can significantly reduce grade 3 or higher chemotherapy-related toxic effects among older adults with cancer receiving chemotherapy and should be included as a part of standard oncology clinical practice.
Abstract
Importance
Although geriatric assessment–driven intervention improves patient-centered outcomes, its influence on chemotherapy-related toxic effects remains unknown.
Objective
To assess whether specific geriatric assessment–driven intervention (GAIN) can reduce chemotherapy-related toxic effects in older adults with cancer.
Design, Setting, and Participants
A randomized clinical trial enrolled 613 participants from a National Cancer Institute–designated cancer center between 2015 and 2019. Patients were 65 years and older with a solid malignant neoplasm, were starting a new chemotherapy regimen, and completed a geriatric assessment. Patients were followed up until chemotherapy completion or 6 months after initiation, whichever occurred first. Data analysis was done by intention-to-treat principle.
Interventions
Patients were randomized (2:1) to either the GAIN (intervention) or standard of care (SOC) arm. In the GAIN arm, a geriatrics-trained multidisciplinary team composed of an oncologist, nurse practitioner, social worker, physical/occupation therapist, nutritionist, and pharmacist reviewed geriatric assessment results and implemented interventions based on prespecified thresholds built into the geriatric assessment’s domains. In the SOC arm, geriatric assessment results were sent to treating oncologists for consideration.
Main Outcomes and Measures
The primary outcome was incidence of grade 3 or higher chemotherapy-related toxic effects (graded using National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.0). Secondary outcomes included advance directive completion, emergency department visits, unplanned hospitalizations, average length of stay, unplanned hospital readmissions, chemotherapy dose modifications, and early discontinuation. Overall survival analysis was performed up to 12 months after chemotherapy initiation.
Results
Among the 605 eligible participants for analysis, median (range) age was 71 (65-91) years, 357 (59.0%) were women, and 432 (71.4%) had stage IV disease. Cancer types included gastrointestinal (202 [33.4%]), breast (136 [22.5%]), lung (97 [16.0%]), genitourinary (91 [15.0%]), gynecologic (54 [8.9%]), and other (25 [4.1%]). Incidence of grade 3 or higher chemotherapy-related toxic effects was 50.5% (95% CI, 45.6% to 55.4%) in the GAIN arm and 60.6% (95% CI, 53.9% to 67.3%) in the SOC arm, resulting in a significant 10.1% reduction (95% CI, −1.5 to −18.2%; P = .02). A significant absolute increase in advance directive completion of 28.4% with GAIN vs 13.3% with SOC (P < .001) was observed. No significant differences were observed in emergency department visits, unplanned hospitalizations, average length of stay, unplanned readmissions, chemotherapy dose modifications or discontinuations, or overall survival.
Conclusions and Relevance
In this randomized clinical trial, integration of multidisciplinary GAIN significantly reduced grade 3 or higher chemotherapy-related toxic effects in older adults with cancer. Implementation of GAIN into oncology clinical practice should be considered among older adults receiving chemotherapy.
Trial Registration
ClinicalTrials.gov Identifier: NCT02517034
Introduction
Cancer is a disease of older adults, with 60% of cancer diagnoses and 70% of cancer deaths occurring in individuals 65 years and older.1 Although age-related deficits, such as functional and cognitive impairments, are common and are associated with outcomes, they are frequently unrecognized or inadequately addressed in oncology practice. The geriatric assessment (GA) is a comprehensive, validated tool that evaluates physical function, comorbidity, nutritional status, polypharmacy, social support, cognition, and psychological status.2 The GA can detect vulnerabilities that are often not captured with standard oncology assessments3 and can better predict chemotherapy-related toxic effects.4,5,6 Therefore, national and international organizations have endorsed including the GA into oncology clinical practice and clinical trials.7,8,9,10,11,12
Identifying geriatric conditions using a GA has helped clinicians develop interventions to maximize overall health in older adults. Geriatric assessment–driven intervention can reduce mortality, hospitalizations, and functional decline in the general older adult population.13,14,15 However, data regarding the effects of GA-driven interventions on clinical outcomes among older adults with cancer are limited.16 We conducted a randomized clinical trial (RCT) to evaluate the efficacy of a specific GA-driven intervention (GAIN) vs standard of care (SOC) on chemotherapy-related toxic effects and other clinical outcomes (advance directive [AD] completion, emergency department [ED] visits, unplanned hospitalizations, average length of stay [ALOS], unplanned hospital readmissions, chemotherapy dose modifications, discontinuations, and survival) among older adults with cancer starting a new chemotherapy regimen. We hypothesized that GAIN would reduce chemotherapy-related toxic effects without compromising survival.
Methods
Study Design
This is an RCT of older adults with a diagnosis of a solid malignant neoplasm starting a new chemotherapy regimen at City of Hope (COH) National Medical Center (Supplement 1). All participants provided written informed consent. The protocol was approved by the COH Institutional Review Board.
Patient Selection
Eligible patients were 65 years and older, with a diagnosis of a solid malignant neoplasm (any stage), starting a new chemotherapy regimen at COH. Any line of cytotoxic chemotherapy, including combination regimens with targeted therapy, was allowed. Patients were fluent in English, Spanish, or Chinese. Trained research staff screened the list of potentially eligible patients and approached them with permission from treating oncologists.
Procedures and Randomization
Before starting chemotherapy, patients completed a baseline GA and the Fulmer SPICES assessment,17,18,19 and the Cancer and Aging Research Group (CARG) chemotherapy toxicity risk score was calculated.4,5 The GA included a patient portion and a health care professional portion (eTable 1 in Supplement 2), consistent with current guidelines.2,7,12,20 The patient portion included self-reported measures of functional status,21,22 comorbidity,22 psychological state,21 social activity/support,23 and nutritional status. Additional demographic variables, including race and ethnicity, were self-reported. Patients selected answers from predefined categories of race (Asian, Black or African American, Native Hawaiian or Other Pacific Islander, Native Indian or Alaska Native, White, unknown) and ethnicity (Hispanic or Latino, non-Hispanic, unknown). This information was collected to identify the racial and ethnic representation of the participants consistent with institutional research procedures. The health care professional portion consisted of physician-rated Karnofsky Performance Status,24 Timed Up and Go test,25 Blessed Orientation-Memory-Concentration test (BOMC, a cognitive screening test),26 weight, height, body mass index, and unintentional weight loss. The GA could be completed on paper or on a touchscreen tablet.27 The Fulmer SPICES assessment evaluates common geriatric syndromes (sleep, eating problems, incontinence, confusion, falls, and skin breakdown) and was administered by a research team member.18,19 Patients were randomized in a 2:1 ratio to either the GAIN or SOC arm, respectively. Blocked randomization with a block size of 12 was used to ensure balanced allocation across arms.28 The randomization list was generated by the study statistician and was password-protected to ensure blinding until eligible patients were ready for allocation. Patients were then treated with chemotherapy under the management of treating oncologists and followed up throughout the duration of their treatment or up to 6 months from the start of chemotherapy (whichever occurred first). Additional follow-up of up to 12 months after chemotherapy initiation was performed only for exploratory survival analysis.
Between August 2015 and February 2019, 613 patients 65 years and older scheduled to begin a new chemotherapy regimen at COH were randomized: 410 to the GAIN arm, and 203 to the SOC arm. Eight patients were found to be ineligible after randomization (7 never started chemotherapy, 1 never received a new chemotherapy regimen) and were excluded from analysis. A total of 605 patients were included in the final analysis, including 5 eligible patients who withdrew consent after randomization (4 from the GAIN arm, 1 from the SOC arm) (Figure 1). No significant differences in withdrawal rates were observed between the GAIN (1.0%) and SOC (0.5%) arms (P = .34). Thirty-one patients (7.7%) from the GAIN arm and 16 (7.9%) from the SOC arm had 6 or more months of chemotherapy, and their follow-up for primary and secondary outcomes was censored at 6 months.
Figure 1. CONSORT Flow Diagram.
GAIN indicates geriatric assessment–driven intervention; SOC, standard of care.
Intervention Delivery
GAIN Arm
Within 2 weeks of study enrollment, each patient’s baseline GA and Fulmer SPICES assessment was reviewed by a geriatrics-trained multidisciplinary team (MDT) composed of an oncologist, nurse practitioner (NP), pharmacist, physical therapist, occupational therapist, social worker and nutritionist (Figure 2). Based on predefined GA thresholds (eTable 2 in Supplement 2) established from literature review and expert consensus, appropriate interventions and referrals were finalized by the MDT. The study NP (L.C.C., J.M., and C.K.) then reviewed the vulnerabilities identified in the GA and the recommended MDT intervention plan and made appropriate referrals while informing both the treating oncologist and the patient. Chemotherapy treatment proceeded at the discretion of treating oncologists. The study NP followed up with the patient to provide additional support (patient education, care coordination, additional specialty referrals) as needed in collaboration with the treating oncology team throughout follow-up.
Figure 2. Geriatric Assessment–Driven Intervention (GAIN) Used in This Study.
Under the guidance of the multidisciplinary team and geriatric nurse practitioner, predefined geriatric assessment thresholds were established and interventions recommended. Please refer to eTable 2 in Supplement 2 for a comprehensive list of intervention recommendations. Body mass index is calculated as weight in kilograms divided by height in meters squared.
SOC Arm
Within 2 weeks of study enrollment, each patient’s baseline GA and Fulmer SPICES assessment was sent to the treating oncologist for self-review without any input from the MDT. Chemotherapy treatment proceeded at the discretion of the treating oncologists, and any specialty referrals were made at their discretion.
For both study arms, if critical issues, such as severe depression (short of suicidality) and/or anxiety (eTable 2 in Supplement 2), were identified through the GA, a social worker referral was made, and the treating oncologist was notified within 24 hours. In cases with an abnormal BOMC score (≥11), indicating potential cognitive impairment, the treating oncologist was also notified. These patient-safety protocols were approved by the COH Institutional Review Board.
Study Outcomes
The primary outcome was the incidence of grade 3 or higher chemotherapy-related toxic effects captured prospectively using the National Cancer Institute Common Terminology Criteria for Adverse Events (version 4.0).29 Secondary outcomes included AD completion, ED visits, unplanned hospitalization(s), ALOS, unplanned hospital readmission rates, chemotherapy dose modifications (reductions and delays), and early discontinuation. To assess these outcomes, each patient’s clinical course was reviewed through medical records. If a patient received emergency care outside COH, patient permission was obtained to review outside medical records. Because of the extensive nature of GAIN, investigators could not be fully blinded to patients’ study arms. To minimize bias, medical record identifiers were removed, and each medical record was reviewed by 2 physicians (D.L. and S.F.D.S.H.) to ensure consistency. Grade 3 or higher chemotherapy-related toxic effects were captured including the specific type of toxic effect along with the category (hematologic or nonhematologic). Additionally, patients consented to be followed up indefinitely to capture late adverse events, including death. We abstracted vital status and date of last contact for participants in both arms through the COH Cancer Registry, with verification of medical records. For survival analysis, patients were observed up to 12 months after initiating chemotherapy. Five patients withdrew consent after randomization, and their medical records were reviewed until the time of withdrawal.
Statistical Analysis
The primary outcome was the incidence of grade 3 or higher chemotherapy-related toxic effects. Based on a 20% difference in toxic effects in the GAIN (30%) compared with the SOC arm (50%), we estimated that a necessary sample size of 600 patients (400 in the GAIN arm and 200 in the SOC arm) to have at least a 90% power to detect this difference, with a 2-sided type I error rate of 0.05. The χ2 and Fisher exact tests as appropriate were used to compare the incidence of grade 3 or higher chemotherapy-related toxic effects, AD completion, ED visits, unplanned hospitalizations, and hospital readmissions between the 2 arms. Kruskal-Wallis test was used to compare ALOS between arms. A 2-sided P value less than .05 was considered statistically significant. Data were analyzed according to intention to treat. Furthermore, we classified patients with grade 3 or higher chemotherapy-related toxic effects into 3 distinct categories: with hematologic toxic effects but without nonhematologic toxic effects (hematologic only); with nonhematologic toxic effects but without hematologic toxic effects (nonhematologic only); and with both hematologic and nonhematologic toxic effects (both hematologic and nonhematologic). Multinomial logistic regression was used to compare the incidence of these 3 categories of toxic effects between both arms.
Survival analysis was performed, with survival time calculated as time from the start of chemotherapy to date of death, date of last contact, or 1 year from chemotherapy initiation, whichever occurred first. Survival probability was estimated using Kaplan-Meier method and log-rank test was used to compare the survival curves between the GAIN and SOC arms. Statistical analysis was performed using SAS, version 9.4 (SAS Institute). This study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline.
Results
Baseline demographic information, disease and treatment characteristics, and GA variables are summarized in Table 1. Median (range) patient age was 71 (65-91) years, and 357 (59.0%) were women. Most patients were White (476 [78.7%]) and non-Hispanic (488 [80.7%]). The most common cancer types were gastrointestinal (202 [33.4%]), breast (135 [22.5%]), lung (97 [16.0%]), genitourinary (91 [15.0%]), and gynecologic (54 [8.9%]); 432 patients (71.4%) had stage IV disease. Overall, patient demographic and disease characteristics were well balanced between the GAIN and SOC arms, with no significant differences observed. In addition, no significant differences between arms were observed in GA domains at baseline or in the calculated CARG chemotherapy toxicity risk scores. There were also no significant differences in treatment characteristics regarding the line or number of chemotherapy agents used, use of concurrent targeted therapy, primary myeloid growth factor prophylaxis, and initial chemotherapy dose reduction between the GAIN and SOC arms. For primary and secondary outcomes, median (interquartile range) duration of follow-up was 85 (50-114) days in the GAIN arm vs 80 (43-133) days in the SOC arm (P = .48). During the study, 3971 potential interventions were identified and recommended for patients in the GAIN arm (mean of 10 per patient), of which 3050 (76.8%) were implemented. Among patients in the SOC arm, 2029 potential interventions were identified by the study team (mean of 10 per patient), of which 254 (12.5%) were implemented by the treating oncologist without receiving recommendations from the MDT. The numbers of recommended and implemented interventions in each specific GA domain by study arm are shown in eTable 3 in Supplement 2.
Table 1. Demographic, Disease, and Treatment Characteristics and Geriatric Assessment Variables.
| Characteristic | No. (%) | ||
|---|---|---|---|
| GAIN (n = 402) | SOC (n = 203) | Overall (n = 605) | |
| Demographic characteristics | |||
| Age, y | |||
| Mean (SD) | 72.0 (5.8) | 72.5 (6.2) | 72.2 (5.9) |
| Median (range) | 71 (65-91) | 72 (65-88) | 71 (65-91) |
| Sex | |||
| Female | 235 (58.5) | 122 (60.1) | 357 (59.0) |
| Male | 167 (41.5) | 81 (39.9) | 248 (41.0) |
| Race | |||
| Asian | 58 (14.4) | 32 (15.8) | 90 (14.9) |
| Black | 27 (6.7) | 9 (4.4) | 36 (6.0) |
| White | 316 (78.6) | 160 (78.8) | 476 (78.7) |
| Othera | 1 (0.2) | 2 (1.0) | 3 (0.5) |
| Ethnicity | |||
| Hispanic | 74 (18.4) | 43 (21.2) | 117 (19.3) |
| Non-Hispanic | 328 (81.6) | 160 (78.8) | 488 (80.7) |
| Disease characteristics | |||
| Cancer type | |||
| Gastrointestinal | 135 (33.6) | 67 (33.0) | 202 (33.4) |
| Breast | 93 (23.1) | 43 (21.2) | 136 (22.5) |
| Lung | 61 (15.2) | 36 (17.7) | 97 (16.0) |
| Genitourinary | 63 (15.7) | 28 (13.8) | 91 (15.0) |
| Gynecologic | 35 (8.7) | 19 (9.4) | 54 (8.9) |
| Other | 15 (3.7) | 10 (4.9) | 25 (4.1) |
| Cancer stage | |||
| I | 26 (6.5) | 5 (2.5) | 31 (5.1) |
| II | 35 (8.7) | 21 (10.3) | 56 (9.3) |
| III | 54 (13.4) | 32 (15.8) | 86 (14.2) |
| IV | 287 (71.4) | 145 (71.4) | 432 (71.4) |
| Treatment characteristics | |||
| Line of chemotherapy | |||
| First | 247 (61.4) | 129 (63.6) | 376 (62.1) |
| Second or beyond | 155 (38.6) | 74 (36.4) | 229 (37.9) |
| No. of chemotherapy agents | |||
| 1 | 129 (32.1) | 66 (32.5) | 195 (32.2) |
| 2 | 256 (63.7) | 132 (65.0) | 388 (64.1) |
| 3 | 17 (4.2) | 5 (2.5) | 22 (3.6) |
| Concurrent targeted therapy | 32 (8.0) | 20 (9.9) | 52 (8.6) |
| Primary myeloid growth factor prophylaxis | 129 (32.1) | 63 (31.0) | 192 (31.7) |
| Initial dose reduction | 148 (36.8) | 87 (42.9) | 235 (38.8) |
| Geriatric assessment variables | |||
| Timed Up and Go, s | |||
| Mean (SD) | 11.3 (3.7) | 11.4 (3.3) | 11.3 (3.6) |
| Median (range) | 10.5 (5.5-40.3) | 10.6 (4.8-30.3) | 10.5 (4.8-40.3) |
| Reported unintentional weight loss | 227 (56.6) | 107 (53.0) | 334 (55.4) |
| Physician-reported KPS | |||
| Median (range) | 80 (50-100) | 80 (40-100) | 80 (40-100) |
| <100% | 291 (72.4) | 158 (77.8) | 449 (76.2) |
| At least 1 fall in last 6 mo | 73 (18.3) | 42 (20.8) | 115 (19.1) |
| ADL score | |||
| Mean (SD) | 59.7 (30.5) | 55.7 (29.7) | 58.3 (30.3) |
| Median (range) | 65 (0-100) | 55 (0-100) | 60 (0-100) |
| IADL score | |||
| Mean (SD) | 12.2 (2.6) | 12.2 (2.3) | 12.2 (2.5) |
| Median (range) | 13 (2-14) | 13 (3-14) | 13 (2-14) |
| No. of comorbidities | |||
| Mean (SD) | 2.4 (1.8) | 2.4 (1.8) | 2.4 (1.8) |
| Median (range) | 2.0 (0-10) | 2.0 (0-9) | 2.0 (0-10) |
| BOMC score | |||
| Mean (SD) | 4.4 (4.3) | 4.8 (4.3) | 4.5 (4.3) |
| Median (range) | 4.0 (0-23) | 4.0 (0-28) | 4.0 (0-28) |
| No. (%) with score ≥11 | 27 (6.7) | 15 (7.4) | 42 (6.9) |
| Social activity | |||
| Mean (SD) | 48.3 (23.7) | 47.9 (23.6) | 48.2 (23.7) |
| Median (range) | 50 (0-100) | 50 (0-91.7) | 50 (0-100) |
| Social support | |||
| Mean (SD) | 84.8 (19.2) | 85.8 (17.7) | 85.1 (18.7) |
| Median (range) | 91.7 (8.3-100) | 91.7 (16.7-100) | 91.7 (8.3-100) |
| Mental Health Inventory-17 | |||
| Mean (SD) | 76.6 (16.7) | 75.7 (17.3) | 76.3 (16.9) |
| Median (range) | 81.2 (3.5-100) | 78.8 (11.8-100) | 80 (3.5-100) |
| BMI | |||
| Mean (SD) | 26.6 (5.5) | 26.7 (6.0) | 26.6 (5.6) |
| Median (range) | 25.8 (16.7-56.6) | 26.1 (16-48.8) | 25.9 (16-56.6) |
| CARG toxicity risk score | |||
| Median (range) | 7 (0-20) | 7 (2-18) | 7 (0-20) |
| Low (0-5) | 111 (27.6) | 50 (24.6) | 161 (26.6) |
| Medium (6-9) | 180 (44.8) | 93 (45.8) | 273 (45.1) |
| High (≥10) | 111 (27.6) | 60 (29.6) | 171 (28.3) |
Abbreviations: ADL, activities of daily living; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); BOMC, Blessed Orientation-Memory-Concentration; CARG, Cancer and Aging Research Group; GAIN, geriatric assessment–driven intervention; IADL, instrumental activities of daily living; KPS, Karnofsky Performance Status; SOC, standard of care.
For race, the “Other” category included a patient who identified as Native American and 2 participants who identified as unknown.
Grade 3 or Higher Chemotherapy-Related Toxic Effects
In the GAIN arm, 203 of 402 patients (50.5%; 95% CI, 45.6% to 55.4%) developed grade 3 or higher chemotherapy-related toxic effects, whereas in the SOC arm, 123 of 203 patients (60.6%; 95% CI, 53.9% to 67.3%) developed grade 3 or higher chemotherapy-related toxic effects. Compared with the SOC arm, the GAIN arm had a statistically significant 10.1% reduction (95% CI, −1.5% to −18.2%; P = .02) in grade 3 or higher chemotherapy-related toxic effects (Table 2). Reductions were observed for hematologic-only toxic effects (8.0% reduction; 95% CI, −1.8% to −14.3%; P = .003) as well as nonhematologic-only toxic effects (8.2% reduction; 95% CI, −1.0% to −15.4%; P = .007) favoring the GAIN arm. No significant difference was observed between treatment arms for both hematologic and nonhematologic toxic effects. Common grade 3 or higher hematologic toxic effects included anemia (112 patients [18.5%]), neutropenia (95 patients [15.7%]), and white blood cell count decrease (56 patients [9.3%]). Common grade 3 or higher nonhematologic toxic effects included infection with normal absolute neutrophil count (103 patients [17.0%]), fatigue (52 patients [8.6%]), and hyponatremia (42 patients [6.9%]).
Table 2. Grade 3 or Higher Chemotherapy-Related Toxic Effects Comparison Between GAIN and SOC Arms.
| Toxic effects | No. (%) | P value | ||
|---|---|---|---|---|
| GAIN (n = 402) | SOC (n = 203) | Total (n = 605) | ||
| Patients with grade 3 or higher chemotherapy-related toxic effects [95% CI] | 203 (50.5) [45.6-55.4] | 123 (60.6) [53.9-67.3] | 326 (53.9) [49.9-57.9] | .02 |
| Hematologic only | 45 (11.2) [8.1-14.3] | 39 (19.2) [13.8-24.6] | 84 (13.9) [11.1-16.6] | .003 |
| Nonhematologic only | 74 (18.4) [14.6-22.2] | 54 (26.6) [20.5-32.7] | 128 (21.2) [17.9-24.4] | .007 |
| Both hematologic and nonhematologic | 84 (20.9) [16.9-24.9] | 30 (14.8) [9.9-19.7] | 114 (18.8) [15.7-22.0] | .64 |
| Type of grade 3 or higher chemotherapy-related toxic effects (with incidence in ≥2% of patients)a | ||||
| Hematologic | ||||
| Anemia | 73 (18.2) | 39 (19.2) | 112 (18.5) | NA |
| Neutropenia | 62 (15.4) | 33 (16.3) | 95 (15.7) | NA |
| White blood cell count decreased | 39 (9.7) | 17 (8.4) | 56 (9.3) | NA |
| Platelet count decreased | 18 (4.5) | 8 (3.9) | 26 (4.3) | NA |
| Febrile neutropenia | 10 (2.5) | 5 (2.5) | 15 (2.5) | NA |
| Nonhematologic | ||||
| Infection with normal ANC | 74 (18.4) | 29 (14.3) | 103 (17.0) | NA |
| Fatigue | 33 (8.2) | 19 (9.4) | 52 (8.6) | NA |
| Hyponatremia | 34 (8.5) | 8 (3.9) | 42 (6.9) | NA |
| Nausea | 14 (3.5) | 13 (6.4) | 27 (4.5) | NA |
| Hypokalemia | 20 (5.0) | 7 (3.5) | 27 (4.5) | NA |
| Dehydration | 15 (3.7) | 11 (5.4) | 26 (4.3) | NA |
| Vomiting | 12 (3.0) | 13 (6.4) | 25 (4.1) | NA |
| Anorexia | 11 (2.7) | 6 (3.0) | 17 (2.8) | NA |
| Generalized muscle weakness | 8 (2.0) | 8 (3.9) | 16 (2.6) | NA |
Abbreviations: ANC, absolute neutrophil count; GAIN, geriatric assessment–driven intervention; NA, not applicable; SOC, standard of care.
Patients could have had more than 1 toxic effect and therefore percentages do not add up to 100%.
Secondary Outcomes
At baseline, the proportion of ADs already on file was similar between the GAIN and SOC arms (46.3% vs 48.8%; P = .56). However, by study end, there were significantly more ADs on file in the GAIN vs the SOC arm (74.6% vs 62.1%; P = .001). Among those who did not have an AD on file at baseline, 28.4% of patients in the GAIN arm vs 13.3% of patients in the SOC arm had signed an AD by the end of follow-up (P < .001) (Table 3). No significant differences were observed between both arms in terms of ED visits, unplanned hospitalizations, ALOS, hospital readmissions, chemotherapy dose modifications, and discontinuation (Table 3).
Table 3. Secondary Outcomes Comparisons Between GAIN and SOC Arms.
| Outcome | No. (%) [95% CI] | P valuea | ||
|---|---|---|---|---|
| GAIN (n = 402) | SOC (n = 203) | Total (n = 605) | ||
| Absolute change in AD statusb | 114 (28.4) [24.0-32.8] | 27 (13.3) [8.6-18.0] | 141 (23.3) [19.9-26.7] | <.001 |
| Emergency department visit | 110 (27.4) [23.0-31.7] | 62 (30.5) [24.2-36.9] | 172 (28.4) [24.8-32.0] | .41 |
| Unplanned hospitalization | 89 (22.1) [18.1-26.2] | 39 (19.2) [13.8-24.6] | 128 (21.2) [17.9-24.4] | .41 |
| Average length of stay, d | ||||
| Mean (SD) | 5.9 (4.2) | 6.8 (5.6) | 6.2 (4.7) | NA |
| Median (range) | 5 (1-23) | 5 (1-26) | 5 (1-26) | .60c |
| Unplanned readmission | 17 (19.1) [10.9-27.3] | 8 (20.5) [7.8-33.2] | 25 (19.5) [12.7-26.4] | .85 |
| Early chemotherapy discontinuation | 216 (53.7) [48.9-58.6] | 118 (58.1) [51.3-64.9] | 334 (55.2) [51.2-59.2] | .30 |
| Chemotherapy dose modificationsd | 218 (54.2) [49.4-59.1] | 95 (46.8) [39.9-53.7] | 313 (51.7) [47.8-55.7] | .08 |
Abbreviations: AD, advance directive; GAIN, geriatric assessment–driven intervention; NA, not applicable; SOC, standard of care.
P values were obtained from χ2 test unless otherwise noted.
Absolute change in AD status reflects the change from no AD at baseline to having an AD at end of primary/secondary outcome follow-up.
P value was obtained from Kruskal-Wallis test.
Dose modifications: reductions or delays.
Survival Analysis
The maximum follow-up time for survival analysis was 12 months after chemotherapy initiation. At that time, 204 patients were deceased, 24 were lost to follow-up, and 377 were still alive. The 6-month and 12-month survival probabilities were 84% and 66% for the GAIN arm and 83% and 64% for the SOC arm, respectively (eFigure in Supplement 2; log-rank P value = .55).
Discussion
To our knowledge, this is the first large RCT evaluating the efficacy of a GA-driven, MDT intervention on serious chemotherapy-related toxic effects and clinical outcomes among older adults with cancer starting a new chemotherapy regimen. GAIN resulted in a significant reduction of more than 10% in grade 3 or higher chemotherapy-related toxic effects when compared with SOC. Additionally, patients randomized to GAIN were significantly more likely to complete an AD. No significant differences were observed in ED visits, unplanned hospitalizations, ALOS, readmissions, chemotherapy dose modifications, discontinuation, or survival between arms.
Reducing chemotherapy-related toxic effects among vulnerable older adults with cancer without compromising treatment efficacy has been challenging for oncologists. Prior smaller studies of older adults have demonstrated mixed results regarding the influence of GA-driven interventions on chemotherapy-related toxic effects. Kalsi et al30 compared older patients with cancer who received GA-driven interventions with historic controls, noting more patients completing the recommended chemotherapy and a trend toward fewer chemotherapy-related toxic effects. In a small randomized pilot study, Magnuson et al31 found no significant changes on toxic effects of a GA-driven intervention vs usual care in adults 70 years and older with stage III or IV solid malignant neoplasms. Prior disease-specific chemotherapy RCTs in older adults, such as the ESOGIA (Elderly Selection on Geriatric Index Assessment) study in non–small cell lung cancer32 or the GO2 trial in gastric cancer,33 have shown that initial dose reductions may lead to a significant reduction in chemotherapy-related toxic effects without compromising survival. In contrast, GAIN demonstrated a statistically significant reduction in grade 3 or higher chemotherapy-related toxic effects without significant differences in treatment characteristics between the GAIN and SOC arms, including the number of chemotherapy agents received, primary myeloid growth factor prophylaxis, and initial dose reductions. Although similar numbers of potential interventions/referrals per patient were identified for both arms, the implementation uptake was much higher in the GAIN arm. This highlights that MDT-based GA-driven interventions can reduce chemotherapy-related toxic effects, without reducing chemotherapy dosing, through improved implementation of recommended interventions. Importantly, 6-month and 12-month survival was similar between arms, providing reassurance that implementing GA-driven interventions mitigates chemotherapy-related toxic effects without compromising efficacy.
Excess medical care use is a challenge among older adults with cancer because of age-associated physiologic decline and multiple comorbidities.34 In the present study, patients in the GAIN arm were significantly more likely to complete an AD. This is most likely owing to having a geriatric-trained NP informing patients on the benefits of AD and helping them overcome identified barriers to AD completion, something shown to improve advance care planning.35,36 In contrast, we did not observe a significant difference in unplanned hospitalizations or ED visits between arms. Several potential risk factors for unplanned hospitalizations among older adults with cancer undergoing chemotherapy have been previously suggested.37 However, designing optimal interventions targeting these risk factors has been challenging, especially for risk factors not easily altered, such as number of comorbid medical conditions, serum albumin levels, or creatinine clearance. Furthermore, interventions aimed at reducing hospital admissions or ED visits often require changes in health care system structure and the standardization of symptom management pathways, which were not components of the present study.38 Future research targeting appropriate interventions to decrease unplanned hospitalizations and ED visits for older adults with cancer is needed to optimize care in this vulnerable population.
This study has several advantages compared with previous studies and ongoing GA-driven intervention trials.39,40 First, baseline patient and treatment characteristics were well balanced with no significant demographic, clinical, or treatment-related differences between the GAIN and SOC arms. This balance limits the likelihood of potential confounding variables influencing the findings and extends the generalizability of the results. Because loss to follow-up is a significant concern when studying a vulnerable group, another strength was the low dropout rate (1%) in the GAIN arm, further supporting the feasibility and acceptability of GA-driven interventions among older adults with cancer. Finally, because patients in both arms completed a baseline GA, the results show the benefit of having a dedicated MDT acting on the results of the GA, compared with only providing those results to the oncology team taking care of the patient. As with other complex health care interventions, it is difficult to pinpoint which specific component was the most important to explain the primary findings. It is likely that, as seen with multicomponent supportive care interventions, a combination of improved symptom management, illness understanding, and enhanced medical and psychological support provided by the MDT is behind the improvements in outcomes.41 Indeed, a high implementation rate of 76.8% of the recommendations was observed in the GAIN arm. Further study is needed to distill which elements of GA-driven care are the most essential.42 Importantly, the fact that 2 other recent RCTs39,40 demonstrated benefits from GA-driven interventions among older adults with cancer receiving treatment shows that including geriatric principles into oncology care leads to better outcomes across differing care models.
Limitations
Study limitations include that it was performed on patients starting a new chemotherapy regimen at a single, tertiary cancer center with a geriatric-focused MDT, potentially limiting the generalizability to settings with differing resources and to other study populations (ie, those receiving targeted therapy or immunotherapy). The time needed to implement a GA in oncology practice is another concern raised by oncology clinicians. However, the mean time needed to complete a GA has been shown to be less than 30 minutes.20,27,43 Recognizing these concerns, we are currently assessing the feasibility of implementing GA-driven interventions via telehealth in community practices with broader study populations. Finally, owing to the nature of the interventions, both patients and treating oncologists were aware of the assignments, which could have potentially influenced care during treatment. However, no differences were found in dose modifications or treatment discontinuation between arms, suggesting that differences in treatment patterns were minimal, supporting the effects of GA-guided interventions on outcomes.
Conclusions
In this randomized clinical trial, integration of GA-driven MDT interventions into oncology care was a clinically meaningful model that led to significant reductions in grade 3 or higher chemotherapy-related toxic effects and improved AD completion among older adults with cancer receiving chemotherapy. Geriatric assessment–driven interventions should be included as a part of standard care for all older adults with cancer.
Trial Protocol
eTable 1. Domains and Measures in the Geriatric Assessment, Descriptive Statistics
eTable 2. Triggers and Intervention Recommendations for Vulnerabilities Identified from the Geriatric Assessment and Fulmer SPICES
eTable 3. Number of Recommended and Implemented Intervention/Referrals By Study Arm
eFigure. Kaplan Meier Curves, Overall Survival, GAIN versus SOC arm
Data Sharing Statement
References
- 1.National Cancer Institute . SEER Cancer Stat Facts: cancer of any site. Accessed August 23, 2021. https://seer.cancer.gov/statfacts/html/all.html
- 2.Hurria A, Gupta S, Zauderer M, et al. Developing a cancer-specific geriatric assessment: a feasibility study. Cancer. 2005;104(9):1998-2005. doi: 10.1002/cncr.21422 [DOI] [PubMed] [Google Scholar]
- 3.Repetto L, Fratino L, Audisio RA, et al. Comprehensive geriatric assessment adds information to Eastern Cooperative Oncology Group performance status in elderly cancer patients: an Italian Group for Geriatric Oncology Study. J Clin Oncol. 2002;20(2):494-502. doi: 10.1200/JCO.2002.20.2.494 [DOI] [PubMed] [Google Scholar]
- 4.Hurria A, Togawa K, Mohile SG, et al. Predicting chemotherapy toxicity in older adults with cancer: a prospective multicenter study. J Clin Oncol. 2011;29(25):3457-3465. doi: 10.1200/JCO.2011.34.7625 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hurria A, Mohile S, Gajra A, et al. Validation of a prediction tool for chemotherapy toxicity in older adults with cancer. J Clin Oncol. 2016;34(20):2366-2371. doi: 10.1200/JCO.2015.65.4327 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Extermann M, Boler I, Reich RR, et al. Predicting the risk of chemotherapy toxicity in older patients: the Chemotherapy Risk Assessment Scale for High-Age Patients (CRASH) score. Cancer. 2012;118(13):3377-3386. doi: 10.1002/cncr.26646 [DOI] [PubMed] [Google Scholar]
- 7.Mohile SG, Dale W, Somerfield MR, et al. Practical assessment and management of vulnerabilities in older patients receiving chemotherapy: ASCO guideline for geriatric oncology. J Clin Oncol. 2018;36(22):2326-2347. doi: 10.1200/JCO.2018.78.8687 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hurria A, Levit LA, Dale W, et al. ; American Society of Clinical Oncology . Improving the evidence base for treating older adults with cancer: American Society of Clinical Oncology statement. J Clin Oncol. 2015;33(32):3826-3833. doi: 10.1200/JCO.2015.63.0319 [DOI] [PubMed] [Google Scholar]
- 9.Levit LA, Singh H, Klepin HD, Hurria A. Expanding the evidence base in geriatric oncology: action items from an FDA-ASCO workshop. J Natl Cancer Inst. 2018;110(11):1163-1170. doi: 10.1093/jnci/djy169 [DOI] [PubMed] [Google Scholar]
- 10.Hurria A, Browner IS, Cohen HJ, et al. Senior adult oncology. J Natl Compr Canc Netw. 2012;10(2):162-209. doi: 10.6004/jnccn.2012.0019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hurria A, Naylor M, Cohen HJ. Improving the quality of cancer care in an aging population: recommendations from an IOM report. JAMA. 2013;310(17):1795-1796. doi: 10.1001/jama.2013.280416 [DOI] [PubMed] [Google Scholar]
- 12.Wildiers H, Heeren P, Puts M, et al. International Society of Geriatric Oncology consensus on geriatric assessment in older patients with cancer. J Clin Oncol. 2014;32(24):2595-2603. doi: 10.1200/JCO.2013.54.8347 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Frese T, Deutsch T, Keyser M, Sandholzer H. In-home preventive comprehensive geriatric assessment (CGA) reduces mortality—a randomized controlled trial. Arch Gerontol Geriatr. 2012;55(3):639-644. doi: 10.1016/j.archger.2012.06.012 [DOI] [PubMed] [Google Scholar]
- 14.Ellis G, Gardner M, Tsiachristas A, et al. Comprehensive geriatric assessment for older adults admitted to hospital. Cochrane Database Syst Rev. 2017;9:CD006211. doi: 10.1002/14651858.CD006211.pub3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Caplan GA, Williams AJ, Daly B, Abraham K. A randomized, controlled trial of comprehensive geriatric assessment and multidisciplinary intervention after discharge of elderly from the emergency department—the DEED II study. J Am Geriatr Soc. 2004;52(9):1417-1423. doi: 10.1111/j.1532-5415.2004.52401.x [DOI] [PubMed] [Google Scholar]
- 16.Li D, Soto-Perez-de-Celis E, Hurria A. Geriatric assessment and tools for predicting treatment toxicity in older adults with cancer. Cancer J. 2017;23(4):206-210. [DOI] [PubMed] [Google Scholar]
- 17.Wallace M, Fulmer T. Fulmer SPICES: an overall assessment tool of older adults. J Gerontol Nurs. 1998;24(12):3. doi: 10.3928/0098-9134-19981201-03 [DOI] [PubMed] [Google Scholar]
- 18.Fulmer T. How to try this: Fulmer SPICES. Am J Nurs. 2007;107(10):40-48. doi: 10.1097/01.NAJ.0000292197.76076.e1 [DOI] [PubMed] [Google Scholar]
- 19.Aronow HU, Borenstein J, Haus F, Braunstein GD, Bolton LB. Validating SPICES as a screening tool for frailty risks among hospitalized older adults. Nurs Res Pract. 2014;2014:846759. doi: 10.1155/2014/846759 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hurria A, Cirrincione CT, Muss HB, et al. Implementing a geriatric assessment in cooperative group clinical cancer trials: CALGB 360401. J Clin Oncol. 2011;29(10):1290-1296. doi: 10.1200/JCO.2010.30.6985 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Stewart AL. Measuring Functioning and Well-being: The Medical Outcomes Study Approach. Duke University Press; 1992. [Google Scholar]
- 22.Fillenbaum GG, Smyer MA. The development, validity, and reliability of the OARS multidimensional functional assessment questionnaire. J Gerontol. 1981;36(4):428-434. doi: 10.1093/geronj/36.4.428 [DOI] [PubMed] [Google Scholar]
- 23.Sherbourne CD, Stewart AL. The MOS social support survey. Soc Sci Med. 1991;32(6):705-714. doi: 10.1016/0277-9536(91)90150-B [DOI] [PubMed] [Google Scholar]
- 24.Karnofsky DA, Burchenal JH. The evaluation of chemotherapeutic agents against neoplastic disease. Cancer Res. 1948;8(8):388-389. 1947 American Association for the Advancement of Science Chemical Research Conference on Cancer abstract. [Google Scholar]
- 25.Podsiadlo D, Richardson S. The timed “Up & Go”: a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 1991;39(2):142-148. doi: 10.1111/j.1532-5415.1991.tb01616.x [DOI] [PubMed] [Google Scholar]
- 26.Kawas C, Karagiozis H, Resau L, Corrada M, Brookmeyer R. Reliability of the blessed telephone information-memory-concentration test. J Geriatr Psychiatry Neurol. 1995;8(4):238-242. doi: 10.1177/089198879500800408 [DOI] [PubMed] [Google Scholar]
- 27.Hurria A, Akiba C, Kim J, et al. Reliability, validity, and feasibility of a computer-based geriatric assessment for older adults with cancer. J Oncol Pract. 2016;12(12):e1025-e1034. doi: 10.1200/JOP.2016.013136 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Altman DG, Bland JM. How to randomise. BMJ. 1999;319(7211):703-704. doi: 10.1136/bmj.319.7211.703 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.National Cancer Institute . Common Terminology Criteria for Adverse Events v4.0. Accessed August 24, 2021. https://ctep.cancer.gov/protocoldevelopment/electronic_applications/ctc.htm
- 30.Kalsi T, Babic-Illman G, Ross PJ, et al. The impact of comprehensive geriatric assessment interventions on tolerance to chemotherapy in older people. Br J Cancer. 2015;112(9):1435-1444. doi: 10.1038/bjc.2015.120 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Magnuson A, Lemelman T, Pandya C, et al. Geriatric assessment with management intervention in older adults with cancer: a randomized pilot study. Support Care Cancer. 2018;26(2):605-613. doi: 10.1007/s00520-017-3874-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Corre R, Greillier L, Le Caër H, et al. Use of a comprehensive geriatric assessment for the management of elderly patients with advanced non-small-cell lung cancer: the phase III randomized ESOGIA-GFPC-GECP 08-02 study. J Clin Oncol. 2016;34(13):1476-1483. doi: 10.1200/JCO.2015.63.5839 [DOI] [PubMed] [Google Scholar]
- 33.Hall PS, Swinson D, Waters JS, et al. Optimizing chemotherapy for frail and elderly patients (pts) with advanced gastroesophageal cancer (aGOAC): the GO2 phase III trial. J Clin Oncol. 2019;37(15)(suppl):4006. [Google Scholar]
- 34.Soto-Perez-de-Celis E, Li D, Yuan Y, Lau YM, Hurria A. Functional versus chronological age: geriatric assessments to guide decision making in older patients with cancer. Lancet Oncol. 2018;19(6):e305-e316. doi: 10.1016/S1470-2045(18)30348-6 [DOI] [PubMed] [Google Scholar]
- 35.Durbin CR, Fish AF, Bachman JA, Smith KV. Systematic review of educational interventions for improving advance directive completion. J Nurs Scholarsh. 2010;42(3):234-241. doi: 10.1111/j.1547-5069.2010.01357.x [DOI] [PubMed] [Google Scholar]
- 36.Lum HD, Sudore RL, Bekelman DB. Advance care planning in the elderly. Med Clin North Am. 2015;99(2):391-403. doi: 10.1016/j.mcna.2014.11.010 [DOI] [PubMed] [Google Scholar]
- 37.Klepin HD, Sun CL, Smith DD, et al. Predictors of unplanned hospitalizations among older adults receiving cancer chemotherapy. JCO Oncol Pract. 2021;17(6):e740-e752. doi: 10.1200/OP.20.00681 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Handley NR, Schuchter LM, Bekelman JE. Best practices for reducing unplanned acute care for patients with cancer. J Oncol Pract. 2018;14(5):306-313. doi: 10.1200/JOP.17.00081 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Mohile SG, Mohamed MR, Culakova E, et al. A geriatric assessment (GA) intervention to reduce treatment toxicity in older patients with advanced cancer: a University of Rochester Cancer Center NCI Community Oncology Research Program cluster randomized clinical trial (CRCT). J Clin Oncol. 2020;38(15)(suppl):12009. [Google Scholar]
- 40.Soo W-K, King M, Pope A, et al. Integrated geriatric assessment and treatment (INTEGERATE) in older people with cancer planned for systemic anticancer therapy. J Clin Oncol. 2020;38(15)(suppl):12011. [Google Scholar]
- 41.Temel JS, Greer JA, Muzikansky A, et al. Early palliative care for patients with metastatic non-small-cell lung cancer. N Engl J Med. 2010;363(8):733-742. doi: 10.1056/NEJMoa1000678 [DOI] [PubMed] [Google Scholar]
- 42.Bauman JR, Temel JS. The integration of early palliative care with oncology care: the time has come for a new tradition. J Natl Compr Canc Netw. 2014;12(12):1763-1771. doi: 10.6004/jnccn.2014.0177 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Hamaker ME, Wildes TM, Rostoft S. Time to stop saying geriatric assessment is too time consuming. J Clin Oncol. 2017;35(25):2871-2874. doi: 10.1200/JCO.2017.72.8170 [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
eTable 1. Domains and Measures in the Geriatric Assessment, Descriptive Statistics
eTable 2. Triggers and Intervention Recommendations for Vulnerabilities Identified from the Geriatric Assessment and Fulmer SPICES
eTable 3. Number of Recommended and Implemented Intervention/Referrals By Study Arm
eFigure. Kaplan Meier Curves, Overall Survival, GAIN versus SOC arm
Data Sharing Statement