If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Prognostic value of baseline functional status measures and geriatric screening in vulnerable older patients with metastatic colorectal cancer receiving palliative chemotherapy – The randomized NORDIC9-study
Department of Oncology, Odense University Hospital, Odense, DenmarkDepartment of Clinical Research, University of Southern Denmark, Odense, DenmarkAcademy of Geriatric Cancer Research (AgeCare), Odense, Denmark
Department of Clinical Research, University of Southern Denmark, Odense, DenmarkAcademy of Geriatric Cancer Research (AgeCare), Odense, DenmarkDepartment of Geriatric Medicine, Odense University Hospital, Odense, Denmark
Department of Clinical Research, University of Southern Denmark, Odense, DenmarkOPEN – Open Patient data Explorative Network, Odense University Hospital, Odense, Denmark
Department of Oncology, Trondheim University Hospital, Trondheim, NorwayDepartment of Clinical and Molecular Medicine, Norwegian University of Science and Technology, Trondheim, Norway
Department of Oncology, Tampere University Hospital and Tampere University, Tampere, FinlandDepartment of Oncology, Helsinki University Hospital, Helsinki, FinlandKarolinska Institutet and Karolinska University Hospital, Stockholm, Sweden
Department of Oncology, Odense University Hospital, Odense, DenmarkDepartment of Clinical Research, University of Southern Denmark, Odense, DenmarkAcademy of Geriatric Cancer Research (AgeCare), Odense, Denmark
Appropriate patient selection based on functional status is crucial when considering older adults for palliative chemotherapy. This pre-planned analysis of the randomized NORDIC9-study explored the prognostic value of four functional status measures regarding progression-free survival (PFS) and overall survival (OS) in vulnerable older patients with metastatic colorectal cancer (mCRC) receiving first-line palliative chemotherapy.
Materials and methods
Patients ≥70 years of age with mCRC not candidates for standard full-dose combination chemotherapy were randomized to receive full-dose S1 or reduced-dose S1 + oxaliplatin. At baseline, functional status was assessed using ECOG performance status (ECOG PS), frailty phenotype, Geriatric 8 (G8), and Vulnerable Elderly Survey-13 (VES-13). Multivariable regression models were applied and C-statistics were estimated.
Results
In total, 160 patients with a median age of 78 years (IQR: 76–81) were included. While in univariate analyses, ECOG PS, frailty phenotype, and VES-13 were statistically significantly associated with differences in OS between subgroups, G8 was not (HR = 1.55, 95%CI: 0.99–2.41, p = 0.050). In multivariable analyses adjusted for age, sex, body mass index, and treatment allocation, we found significant differences between subgroups for all applied tools and with C-statistics in the moderate range for ECOG PS and VES-13.
Concerning PFS, statistically significant differences were observed between subgroups of ECOG PS, G8, and VES-13 both in uni- and multivariable analyses, but not for frailty phenotype.
Discussion
In this Nordic cohort of vulnerable older patients with mCRC, baseline ECOG PS, frailty phenotype, G8, and VES-13 showed prognostic value regarding overall survival, and moderate predictive value of models based on ECOG PS and VES-13 was demonstrated.
]. The evidence cannot be directly extrapolated to most patients treated in clinical practice who often have comorbidities, impaired organ function, geriatric syndromes (fall tendency, incontinence, osteoporosis, cognitive impairment, polypharmacy), and are at risk of developing frailty. Frailty is a common clinical syndrome in older adults resulting in increased vulnerability to stressors [
]. Frail patients are at significant risk of being undertreated, having shorter survival, experiencing more toxicities, and worse quality-of-life (QoL) [
What every oncologist should know about geriatric assessment for older patients with cancer: young International Society of Geriatric Oncology Position Paper.
It has repeatedly been questioned how older vulnerable patients with metastatic colorectal cancer (mCRC) should optimally be treated. Yet, few RCTs have investigated different treatment approaches in this setting [
Geriatric analysis from PRODIGE 20 randomized phase II trial evaluating bevacizumab + chemotherapy versus chemotherapy alone in older patients with untreated metastatic colorectal cancer.
Reduced-dose combination chemotherapy (S-1 plus oxaliplatin) versus full-dose monotherapy (S-1) in older vulnerable patients with metastatic colorectal cancer (NORDIC9): a randomised, open-label phase 2 trial.
]. Neither consensus nor uniform practice has been achieved so far. The NORDIC9-study included 160 patients ≥70 years treated with either full-dose monotherapy (S1) or reduced-dose combination chemotherapy (S1 + oxaliplatin) and established a potential new standard of care in this population [
Reduced-dose combination chemotherapy (S-1 plus oxaliplatin) versus full-dose monotherapy (S-1) in older vulnerable patients with metastatic colorectal cancer (NORDIC9): a randomised, open-label phase 2 trial.
]. Reduced-dose doublet chemotherapy resulted in significantly prolonged progression-free survival (PFS), less toxicity, and preservation of physical functioning and QoL [
Reduced-dose combination chemotherapy (S-1 plus oxaliplatin) versus full-dose monotherapy (S-1) in older vulnerable patients with metastatic colorectal cancer (NORDIC9): a randomised, open-label phase 2 trial.
]. Oncologists usually use the Eastern Cooperative Oncology Group Performance Status (ECOG PS) in clinical practice, though it has poor correlation with the CGA, and its utility questioned in older adults with cancer [
]. The frailty phenotype is a well-established model in geriatric medicine, however, containing measurements not used in oncology routinely, like handgrip strength and gait speed [
]. A compromise might be the application of brief, simple geriatric screening tools fitting the daily oncology practice such as Geriatric 8 (G8) and Vulnerable Elderly Survey-13 (VES-13) [
Geriatric screening in older adults with cancer - a young International society of geriatric oncology and nursing & allied health interest group initiative.
]. Screening tools do not require competences in geriatric medicine and are freely available in both paper form, on-line, and as mobile application (Oncoassist®). Furthermore, they can be completed by the patient or caregivers in 5–10 min, and provide information about survival, functional decline, and chemotherapy toxicity [
The overarching aim of this current pre-planned analysis was to assess the prognostic performance of the different functional status measures conducted at baseline regarding OS and PFS.
2. Patients and Methods
2.1 Study design and Participants
The NORDIC9-study, a randomized multi-center study included patients ≥70 years with mCRC who were not candidates for standard full-dose combination chemotherapy (EudraCT reg.no. 2014–000394-39). The detailed study protocol, the primary and several secondary endpoints have been published [
Reduced-dose combination chemotherapy (S-1 plus oxaliplatin) versus full-dose monotherapy (S-1) in older vulnerable patients with metastatic colorectal cancer (NORDIC9): a randomised, open-label phase 2 trial.
Randomized study comparing full dose monotherapy (S-1 followed by irinotecan) and reduced dose combination therapy (S-1/oxaliplatin followed by S-1/irinotecan) as initial therapy for older patients with metastatic colorectal cancer: NORDIC 9.
In brief, participants were randomly allocated (1:1) to treatment with either full-dose S1 monotherapy (30 mg/m2 orally twice-daily on days 1–14, every three weeks (q3w)) or with reduced-dose SOx (S1, 20 mg/m2 orally twice daily + oxaliplatin 100 mg/m2 intravenously on day 1, q3w), the addition of bevacizumab (7.5 mg/kg intravenously, q3w) was optional. Participants were treated until disease progression, unacceptable toxicity, or patient wish.
2.3 Key Variables of Interest
Four measurements of functional status were registered at baseline: ECOG PS, frailty phenotype, G8, and VES-13.
2.4 ECOG PS
The most commonly applied physician-reported measurement of functional status in oncology [
]. The physician chooses the one statement appropriately describing the patient's level of physical activity. Patients with ECOG PS ≥3 are usually not considered as candidates for anti-neoplastic treatment.
2.5 Frailty Phenotype
Frailty phenotype covers five domains: weight loss, exhaustion, low physical activity, weakness, and slow gait. Score ranges from zero to five. A score of 0–2 is categorized as non-frail, and a score of ≥3 as frail [
Domains were derived from our collected data as following: for weight loss, we used the patients' self-reported weight loss (>5% during the two months prior to inclusion). Exhaustion was derived from the patient-reported European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core-30 version 3.0 (EORTC QLQ-C30) fatigue domain [
The European organization for research and treatment of cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology.
]. The questionnaire asks about the last one-week period and applies a four-point response format ranging from “not at all” to “very much”. The raw score was linearly transformed to a score between 0 and 100 according to the EORTC scoring manual [
]. The higher value means larger symptom burden. We used the recommended threshold for clinical importance (TCI) predefined by the EORTC QoL expert panel (39 point) as cut-off (fatigue vs no fatigue) [
]. Low physical activity was also derived from EORTC QLQ-C30 using the physical functioning domain. We applied the same scoring procedure, as described above; though here, a higher score indicates better functioning. We applied the predefined TCI at 83 points [
]. Weakness was defined as reduced handgrip strength measured by hand-held dynamometer; the lowest sex-adjusted 20% percentile was considered weak. Slow gait was considered as the lowest 20% percentile of the Timed Up and Go test.
2.6 G8
G8 is an eight-item questionnaire addressing geriatric domains with a maximum score of 17 [
]. It focuses on activities of daily living (ADL) and the instrumental activities of daily living (IADL). The maximum score is 10 and a score ≥ 3 indicates frailty.
2.7.1 Outcomes
The prognostic performance of ECOG PS, frailty phenotype, G8, and VES-13 according to OS and PFS in the NORDIC9-study.
2.7.2 Sample size
For this analysis, no formal sample size calculation was performed; the current sample follows the sample size calculated for the PFS endpoint of the NORDIC9-study (intention-to-treat (ITT) population: n = 160) [
Reduced-dose combination chemotherapy (S-1 plus oxaliplatin) versus full-dose monotherapy (S-1) in older vulnerable patients with metastatic colorectal cancer (NORDIC9): a randomised, open-label phase 2 trial.
For baseline demographical and clinical characteristics, we applied descriptive statistics. Depending on the number of observations, for categorical variables chi-squared-test or Fischer's exact test was used, for continuous numerical variables the Wilcoxon Mann-Whitney test was applied.
2.7.4 Survival analyses
The starting time point of follow-up for all included patients was the time point of inclusion when the patient signed the informed consent form. For OS and PFS, survival curves were estimated by the Kaplan-Meier method, the comparisons between sub-groups were performed by log-rank test. Hazard ratios (HR) and corresponding 95% confidence intervals (95%CI) were estimated by Cox proportional hazard regression and the proportional hazards assumptions were tested by Schoenfield residuals. All physical functioning measurements were analyzed separately.
2.7.5 Multivariable analyses
We constructed a multivariable regression model applying Cox proportional hazards regression for the survival outcomes. Clinically relevant covariates were identified and univariate analyses were conducted assessing the possible association between baseline characteristics and measurements of functional status. The co-variable was included in the model if the p-value was <0.1 or a co-variable was considered clinically relevant. The model hence was adjusted for age, sex, body mass index (BMI), and treatment allocation. We balanced the number of co-variables according to the number of observations to avoid over-fitting. The four measurements of functional status were tested one by one in the models for both OS and PFS; adding the co-variables one by one to the models allowed us to conduct sensitivity analyses. In these analyses, we also included the different functional status measures one by one being able to evaluate whether they improve the models.
Comparing the different models of functional status measurements beside HRs and statistical significance, we also applied C-statistics and calculated Harrell's C (area under the curve (AUC)) values from Cox proportional hazards regression models demonstrating the diagnostic ability of ECOG PS, frailty phenotype, G8, and VES-13. We included only the complete cases when estimating C-statistics to be able to provide a sensitivity analysis, hence, comparable Harrell's C values.
Two-sided p-values ≤0.05 were considered statistically significant and estimates were reported with 95%CI.
2.7.6 Missing data
Only 2–6% of observations were missing for variables in the dataset, so we concluded that excluding those observations from analysis was reasonable.
2.7.7 Software
We performed data analysis in STATA v17 (StataCorp LLC, College Station, TX, USA).
3. Results
3.1 Patient Population
Between March 2015 and October 2017, 160 patients were included in the NORDIC9-study and available for analysis [
Randomized study comparing full dose monotherapy (S-1 followed by irinotecan) and reduced dose combination therapy (S-1/oxaliplatin followed by S-1/irinotecan) as initial therapy for older patients with metastatic colorectal cancer: NORDIC 9.
]. The median follow-up was 23.8 months (interquartile range (IQR): 18.8–30.9). The participant flow is presented as a CONSORT 2010 diagram (Fig. 1). The inclusion was ended when the required number of patients was achieved.
The median age was 78 years (IQR: 76–81). Patient, disease, and clinical characteristics were well balanced between the treatment arms (Table 1).
Table 1Demographic and baseline clinical characteristics of the intention-to-treat population in the NORDIC9-study and statistical comparison of the treatment arms.
Baseline characteristics Data presented as median (IQR) or n (%) as appropriate.
NORDIC9-study Intention-to-treat population n = 160
NORDIC9-study treatment arms
Comparison of Arm A and B
p-value
Full-dose monotherapy (S1) Arm A n = 83
Reduced-dose doublet (SOx) Arm B n = 77
Age
0.5631
Median age in years (IQR)
78 (75–81)
78 (76–81)
78 (75–80)
70–74 years
35 (22%)
16 (19%)
19 (25%)
75–79 years
74 (46%)
38 (46%)
36 (47%)
≥ 80 years
51 (32%)
29 (35%)
22 (29%)
Sex
0.8840
Female
78 (49%)
40 (48%)
38 (49%)
Male
82 (51%)
43 (52%)
39 (51%)
Location of primary tumor
0.9930
Left sided
97 (61%)
50 (60%)
47 (61%)
Right sided
62 (39%)
32 (39%)
30 (39%)
Surgery for primary tumor
0.7000
No
69 (43%)
37 (45%)
32 (42%)
Yes
91 (57%)
46 (55%)
45 (58%)
Prior adjuvant chemotherapy
0.2250
Yes
29 (18%)
18 (22%)
11 (14%)
No
131 (82%)
65 (78%)
66 (86%)
Presentation at diagnosis
0.1750
Synchronous
96 (60%)
54 (65%)
42 (55%)
Metachronous
64 (40%)
29 (35%)
35 (45%)
Number of metastatic sites
0.9630
1–2
127 (79%)
66 (80%)
61 (79%)
≥3
33 (21%)
17 (20%)
16 (21%)
Sites of metastatic disease
Liver
102 (64%)
58 (70%)
44 (57%)
0.0940
Lung
65 (41%)
34 (41%)
31 (40%)
0.9280
Lymph nodes
69 (43%)
41 (52%)
28 (39%)
0.1390
Peritoneum
40 (25%)
12 (14%)
28 (36%)
0.0010
Bone
6 (4%)
3 (4%)
3 (4%)
0.7750
Other
25 (16%)
12 (14%)
13 (17%)
0.8830
Self-reported weight-loss > 5% within the last 2 months
In univariate analyses, both ECOG PS 1 and 2 resulted in statistically significantly shorter OS compared to ECOG PS 0: 21.4 months for ECOG PS 0, 13.1 months for ECOG PS 1, and 10.3 months for ECOG PS 2, respectively (Table 2, Fig. 2). In addition, a significant difference was found between ECOG PS 0 and 2 regarding PFS (8.3 (95%CI: 5.9–10.2) vs 3.9 months (95%CI 3.1–5.5), HR = 1.96 (1.24–3.08), p = 0.004) (Table 2,Fig. 3).
Table 2Summary of progression-free survival and overall survival in univariate models according to the Eastern Cooperative Oncology Group Performance Status (ECOG PS), Frailty phenotype, Geriatric 8, and Vulnerable Elderly Survey-13.
Progression-Free and Overall Survival according to functional status measurements Univariate analyses
Fig. 2Univariate Kaplan-Meier survival estimates of overall survival (OS) according to Eastern Cooperative Oncology Group Performance Status (ECOG PS), Frailty phenotype, Geriatric 8 (G8), and Vulnerable Elderly Survey-13 (VES-13).
Fig. 3Univariate Kaplan-Meier survival estimates of progression-free survival (PFS) according to Eastern Cooperative Oncology Group Performance Status (ECOG PS), Frailty phenotype, Geriatric 8 (G8), and Vulnerable Elderly Survey-13 (VES-13), follow-up time is censored at 18-month.
In univariate analyses, patient stratification based on frailty phenotype found a significant difference in OS between the non-frail and frail group: 14.2 (95%CI: 11.3–19.6) vs 12.9 months (95%CI: 5.6–14.4), HR = 1.63, (95%CI: 1.04–2.55), p = 0.031) (Fig. 2,Table 2.). The significance was maintained in multivariable analysis adjusted for age, sex, BMI, and treatment allocation (HR: 1.68 (95%CI: 1.07–2.65), p = 0.025).
Regarding PFS, no statistically significant differences between sub-groups of frailty phenotype were observed in univariate analyses (Fig. 3,Table 2, and Table 3).
Table 3The summary of progression-free survival and overall survival applying the Eastern Cooperative Oncology Group Performance Status (ECOG PS), Frailty phenotype, Geriatric 8, and Vulnerable Elderly Survey-13, using the main effects multivariable models.
Progression-free and overall survival according to functional status measurements Multivariable analyses
Physical functioning measurements
n
Progression-free survival
Overall survival
HR
95% CI
p-value
Harrell's C (95% CI)
HR
95% CI
p-value
Harrell's C (95% CI)
ECOG PS
0
53
1.00
0.63 0.58–0.68
1.00
0.64 (0.58–0.70)
1
75
1.55
1.05–2.29
0.028
1.99
1.26–3.16
0.003
2
32
2.47
1.48–4.12
0.001
3.32
1.89–5.83
<0.001
Frailty phenotype
Non-frail
131
1.00
0.176
0.58 (0.52–0.63)
1.00
0.025
0.58 (0.52–0.64)
Frail
29
1.35
0.86–2.08
1.68
1.07–2.65
Geriatric 8
>14
44
1.00
0.011
0.57 (0.52–0.62)
1.00
0.038
0.59 (0.53–0.65)
≤14
110
1.65
1.12–2.42
1.62
1.03–2.55
Vulnerable Elderly Survey-13
0–2
113
1.00
0.003
0.60 (0.55–0.65)
1.00
<0.001
0.61 (0.54–0.67)
≥3
36
1.81
1.22–2.70
2.29
1.48–3.56
The models for ECOG PS, frailty phenotype, and Vulnerable Elderly Survey-13 were adjusted for age, sex, body mass index, and treatment allocation. The model for Geriatric 8 was adjusted for age, sex, and treatment allocation but not body mass index to avoid multicollinearity.
Applying G8, OS tended to be statistically significant in favor of those with a score > 14: 18.6 (95%CI: 12.3–27.6) vs 11.5 (95% CI: 10.3–14.4) months (HR = 1.55 (95%CI: 0.99–2.41), p = 0.050). The PFS difference was 8.3 vs 5.3 months (HR = 1.63, p = 0.009).
Stratification based on VES-13 resulted in statistically significant OS and PFS differences (OS: 15.9 (95%CI: 12.3–21.2) vs 6.5 months (95% CI: 5.3–13.4), HR = 2.12 (95%CI: 1.38–3.24), p = 0.001), PFS: 6.5 (95%CI: 5.5–8.1) vs 3.4 months (95%CI: 2.3–4.6), HR = 1.86 (95%CI: 1.27–2.74), p = 0.002).
We created a table (Supplementary Table 1) showing failure rates in different time points during follow-up according to survival outcomes in subgroups of functional status measures and geriatric screening tools; our data are consistent regarding OS and PFS in the subgroups, such as for OS at 1-, 2-, and 3-year follow-up.
Applying multivariable analyses, we found statistically significant differences in all models for OS; the highest HRs were observed between ECOG PS (for ECOG PS: 1: HR: 1.99 (95%CI: 1.26–3.16), p = 0.003; for ECOG PS: 2: HR: 3.32 (95%CI: 1.89–5.83), p < 0.001) and VES-13 sub-groups (for VES-13 ≥ 3: HR: 2.29 (95%CI: 1.48–3.56), p < 0.001) (Table 3). While C-statistics showed a moderate prediction for these two models with an AUC above 0.6; for frailty phenotype and G8 the predictive ability was below 0.60, (0.58 and 0.59, respectively) (Table 3).
3.2 Sensitivity analyses
Our sensitivity analysis applying stepwise addition of variables showed improved sensitivity when age, sex, BMI, and treatment allocation were included in the models, though, further adjustments for ECOG PS, frailty phenotype, G8, and VES-13 did not enhance the sensitivity (Supplementary Table 2).
4. Discussion
4.1 Summary of the Results
We found that functional status by ECOG PS, frailty phenotype, G8, and VES-13 was significantly associated with OS in multivariable analyses, thus, all applied tools demonstrated prognostic value. Moderate prediction of the models ECOG PS and VES-13 was shown in the NORDIC9 cohort.
4.2 Perspective/Clinical Context
Using the well-established ECOG PS in daily oncology practice provides important prognostic information. The addition of geriatric screening tools may add important details on the challenges older vulnerable patients face, such as weight-loss, mobility issues, medication use, cognitive issues, and help needed for ADL and IADL (housekeeping, assisting with self-care, shopping). This information may contribute to more proper prognostic understanding, help to explore patient and caregiver preferences, provide more details for shared decision-making, guide personalized interventions, and improve QoL [
Perspectives on functional status in older adults with cancer: an interprofessional report from the International Society of Geriatric Oncology (SIOG) nursing and allied health interest group and young SIOG.
A systematic review assessing the sensitivity and specificity of seven geriatric screening tools in older patients with cancer, including frailty phenotype, G8, and VES-13 found the lowest median sensitivity (31%) for frailty phenotype with a high specificity(91%), while the sensitivity of G8 and VES-13 were 87% and 68%, and the specificity 61% and 78%, respectively [
]. Despite ECOG PS provides a shallow assessment of physical function, in our cohort it demonstrated comparable prognostic approximation to VES-13. A possible explanation may be that both tools assess ADL, though, ECOG PS only is considered as a shallow description of these activities. Of note, ECOG PS is the tool oncologists use most often and are most familiar with, patient stratification based on ECOG PS, thus, reflects to their treatment pattern, habits, and experience.
Combining screening tools, such as G8 and VES-13 showed increased sensitivity and specificity in terms of identifying frailty [
Reduced-dose combination chemotherapy (S-1 plus oxaliplatin) versus full-dose monotherapy (S-1) in older vulnerable patients with metastatic colorectal cancer (NORDIC9): a randomised, open-label phase 2 trial.
Combination of the G-8 screening tool and hand-grip strength to predict long-term overall survival in non-small cell lung cancer patients undergoing stereotactic body radiotherapy.
], which could be explained by different populations, study design, and methodology.
Of note, the details obtained by CGA may provide more precise prognostic and predictive information than screening tools, ECOG PS, and frailty phenotype. CGA may predict chemotherapy toxicity, surgical complications, mortality, and QoL as highlighted in recent publications, hence, CGA should be considered as the standard of care in the multidisciplinary management of older patients with cancer [
Geriatric impairments are prevalent and predictive of survival in older patients with cancer receiving radiotherapy: a prospective observational study.
4.4 Interpretation, Explanation, and Methodological Considerations
In the NORDIC9-study, all applied tools measuring functional status provided prognostic information on OS; however, considering the AUC values, only ECOG PS and VES-13 showed moderate prediction. It is though important to note that patient stratification was primarily based on ECOG PS at inclusion; G8 and VES-13 were applied and frailty phenotype was estimated after the patients entered the study. This might have influenced the outcomes.
Despite the statistically significant OS difference, the prediction of frailty phenotype was low. A possible explanation is that frailty phenotype has been developed and applied in geriatric medicine, thus, patients considered frail according to frailty phenotype are “more“frail than those usually considered for anti-neoplastic treatment. The patients usually managed by geriatricians are therefore not included in clinical trials and often receive best supportive care exclusively. Moreover, given the design of our study and the collected dataset, we could not apply the original measurements used by Fried [
]. However, we considered our surrogate measures as an appropriate approximation of the original measures.
We applied derived variables for fatigue and physical functioning, using the EORTC QLQ-C30 questionnaire and the TCIs recommended by the EORTC expert panel. These TCI values were developed in younger and less comorbid cohorts of patients with different types of malignancies, not in vulnerable older patients with mCRC. Equivalent TCIs are not available in these patients yet. We also considered applying thresholds available from a general healthy older Danish population, though it may not reflect our vulnerable older population appropriately. Specific TCIs developed in this particular population might have improved our frailty phenotype model.
We evaluated four different functional status measurements in a prospective randomized multicenter setting in a homogenous cohort of vulnerable older patients with mCRC. To the best of our knowledge, the utility of frailty phenotype has not previously been assessed in this setting.
Our study has limitations. As mentioned above, we used surrogate values derived from EORTC QLQ-C30; using the original frailty phenotype criteria might have shown different results. The application of TCI developed in a specific cohort of vulnerable older patients with mCRC might have contributed to improved sensitivity and specificity of our frailty phenotype model.
5. Conclusions
We established that ECOG PS, frailty phenotype, G8, and VES-13 were associated with OS and moderate prediction of ECOG PS and VES-13 was demonstrated in vulnerable older patients with mCRC receiving palliative chemotherapy. ECOG PS is already an established tool in daily oncology practice. The addition of VES-13 systematically in daily oncology practice may add important information about IADL and lead to personalized interventions, thus potentially improving the decision-making and outcomes in this population.
Funding
This investigator-initiated study was partly funded by Taiho Pharmaceuticals, Nordic Group, The Danish Cancer Society (grant number: R269-A15859), Academy of Geriatric Cancer Research (AgeCare), The Swedish Cancer Society and the Region of Southern Denmark.
Institutional review board statement
The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Regional Scientific Ethical Committee (project ID: S-20140020, date of approval: 08.05.2014), the Danish Health and Medicines Authority (journal nr.: 2014023387), and the Danish Data Protection Agency (number of approval: 2008-58-0035). EudraCT-number: 2014–000394-39. Finland: Operatiivinen eettinen toimikunta at Helsinki University Hospital (357/13/03/02/14, the date of approval: 21.10.2015). Norway: Regional Committee for Medical and Health Research Ethics, REC West, Norway (2014/598, the date of approval: 22.05.2014). Sweden: Regionala Etikprovningsnämnden, Uppsala (2014/288, the date of approval: 19.11.2014).
Informed consent statement
Informed consent was obtained from all subjects involved in the study.
The authors have no competing interest to declare. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
Acknowledgments
The investigators sincerely thank the patients, their family members and caregivers, all involved health-care providers, and everyone who has contributed to this work.
What every oncologist should know about geriatric assessment for older patients with cancer: young International Society of Geriatric Oncology Position Paper.
Geriatric analysis from PRODIGE 20 randomized phase II trial evaluating bevacizumab + chemotherapy versus chemotherapy alone in older patients with untreated metastatic colorectal cancer.
Reduced-dose combination chemotherapy (S-1 plus oxaliplatin) versus full-dose monotherapy (S-1) in older vulnerable patients with metastatic colorectal cancer (NORDIC9): a randomised, open-label phase 2 trial.
Geriatric screening in older adults with cancer - a young International society of geriatric oncology and nursing & allied health interest group initiative.
Randomized study comparing full dose monotherapy (S-1 followed by irinotecan) and reduced dose combination therapy (S-1/oxaliplatin followed by S-1/irinotecan) as initial therapy for older patients with metastatic colorectal cancer: NORDIC 9.
The European organization for research and treatment of cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology.
Perspectives on functional status in older adults with cancer: an interprofessional report from the International Society of Geriatric Oncology (SIOG) nursing and allied health interest group and young SIOG.
Geriatric impairments are prevalent and predictive of survival in older patients with cancer receiving radiotherapy: a prospective observational study.
Combination of the G-8 screening tool and hand-grip strength to predict long-term overall survival in non-small cell lung cancer patients undergoing stereotactic body radiotherapy.
00530-6&id=gr1.jpg){kind=link}
00530-6&id=gr2.jpg){kind=link}
00530-6&id=gr3.jpg){kind=link}