|

 

Early palliative care for patients with metastatic non–small-cell lung cancer

Early Palliative Care for Patients with Metastatic Non–Small-Cell Lung Cancer Jennifer S. Temel, M.D., Joseph A. Greer, Ph.D., Alona Muzikansky, M.A., Emily R. Gallagher, R.N., Sonal Admane, M.B., B.S., M.P.H., Vicki A. Jackson, M.D., M.P.H., Constance M. Dahlin, A.P.N., Craig D. Blinderman, M.D., Juliet Jacobsen, M.D., William F. Pirl, M.D., M.P.H., J. Andrew Billings, M.D., and Thomas J. Lynch, M.D.
Background
Patients with metastatic non–small-cell lung cancer have a substantial symptom From Massachusetts General Hospital,
Boston (J.S.T., J.A.G., A.M., E.R.G., V.A.J., burden and may receive aggressive care at the end of life. We examined the effect C.M.D., J.J., W.F.P., J.A.B.); the State Uni- of introducing palliative care early after diagnosis on patient-reported outcomes versity of New York, Buffalo (S.A.); Adult and end-of-life care among ambulatory patients with newly diagnosed disease.
Palliative Medicine, Department of Anes-thesiology, Columbia University Medical Center, New York (C.D.B.); and Yale Uni- versity, New Haven, CT (T.J.L.). Address We randomly assigned patients with newly diagnosed metastatic non–small-cell reprint requests to Dr. Temel at Massa- chusetts General Hospital, 55 Fruit St., lung cancer to receive either early palliative care integrated with standard onco- Yawkey 7B, Boston, MA 02114, or at logic care or standard oncologic care alone. Quality of life and mood were assessed [email protected].
at baseline and at 12 weeks with the use of the Functional Assessment of Cancer N Engl J Med 2010;363:733-42.
Therapy–Lung (FACT-L) scale and the Hospital Anxiety and Depression Scale, re- Copyright 2010 Massachusetts Medical Society. spectively. The primary outcome was the change in the quality of life at 12 weeks. Data on end-of-life care were collected from electronic medical records.
Results
Of the 151 patients who underwent randomization, 27 died by 12 weeks and 107
(86% of the remaining patients) completed assessments. Patients assigned to early palliative care had a better quality of life than did patients assigned to standard care (mean score on the FACT-L scale [in which scores range from 0 to 136, with higher scores indicating better quality of life], 98.0 vs. 91.5; P = 0.03). In addition, fewer patients in the palliative care group than in the standard care group had depressive symptoms (16% vs. 38%, P = 0.01). Despite the fact that fewer patients in the early palliative care group than in the standard care group received aggressive end-of-life care (33% vs. 54%, P = 0.05), median survival was longer among patients receiving early palliative care (11.6 months vs. 8.9 months, P = 0.02).
Conclusions
Among patients with metastatic non–small-cell lung cancer, early palliative care led
to significant improvements in both quality of life and mood. As compared with patients receiving standard care, patients receiving early palliative care had less aggressive care at the end of life but longer survival. (Funded by an American Society of Clinical Oncology Career Development Award and philanthropic gifts; ClinicalTrials.gov number, NCT01038271.) n engl j med 363;8 nejm.org august 19, 2010 The New England Journal of Medicine Downloaded from www.nejm.org on September 2, 2010. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. The quality of care and the use of static non–small-cell lung cancer in a nonblind- medical services for seriously ill patients ed, randomized, controlled trial of early palliative are key elements in the ongoing debate care integrated with standard oncologic care, as over reform of the U.S. health care system.1 On- compared with standard oncologic care alone. cologic care is central to this debate, largely be- The study was performed at Massachusetts Gen- cause anticancer treatments are often intensive eral Hospital in Boston. Eligible patients were and costly.2 Comprehensive oncologic services for enrolled within 8 weeks after diagnosis and were patients with metastatic disease would ideally randomly assigned to one of the two groups in a improve the patients' quality of life and facilitate 1:1 ratio without stratification. Patients who were the efficient allocation of medical resources. Pal- assigned to early palliative care met with a mem- liative care, with its focus on management of ber of the palliative care team, which consisted symptoms, psychosocial support, and assistance of board-certified palliative care physicians and with decision making, has the potential to im- advanced-practice nurses, within 3 weeks after prove the quality of care and reduce the use of enrollment and at least monthly thereafter in the medical services.3,4 However, palliative care has outpatient setting until death. Additional visits traditionally been delivered late in the course of with the palliative care service were scheduled at disease to patients who are hospitalized in spe- the discretion of the patient, oncologist, or pallia- cialized inpatient units or as a consultative ser- tive care provider.
vice for patients with uncontrolled symptoms.5,6 General guidelines for the palliative care vis- Previous studies have suggested that late refer- its in the ambulatory setting were adapted from rals to palliative care are inadequate to alter the the National Consensus Project for Quality Pallia- quality and delivery of care provided to patients tive Care and were included in the study protocol.14 with cancer.7,8 To have a meaningful effect on Using a template in the electronic medical re- patients' quality of life and end-of-life care, pallia- cord, palliative care clinicians documented the tive care services must be provided earlier in the care they provided according to these guidelines course of the disease.
(see Table 1 in the Supplementary Appendix, avail- Metastatic non–small-cell lung cancer, the able with the full text of this article at NEJM.org). leading cause of death from cancer worldwide,9 Specific attention was paid to assessing physical is a debilitating disease that results in a high and psychosocial symptoms, establishing goals burden of symptoms and poor quality of life; the of care, assisting with decision making regard- estimated prognosis after the diagnosis has been ing treatment, and coordinating care on the established is less than 1 year.10-12 We previous- basis of the individual needs of the patient.14,15 ly found that introducing palliative care shortly Patients who were randomly assigned to standard after diagnosis was feasible and acceptable among care were not scheduled to meet with the pallia- outpatients with metastatic non–small-cell lung tive care service unless a meeting was requested cancer.13 The goal of the current study was to by the patient, the family, or the oncologist; those examine the effect of early palliative care inte- who were referred to the service did not cross grated with standard oncologic care on patient- over to the palliative care group or follow the reported outcomes, the use of health services, specified palliative care protocol. All the partici- and the quality of end-of-life care among patients pants continued to receive routine oncologic care with metastatic non–small-cell lung cancer. We throughout the study period. Before enrollment hypothesized that patients who received early in the study was initiated, the protocol was ap- palliative care in the ambulatory care setting, as proved by the Dana Farber/Partners CancerCare compared with patients who received standard institutional review board. All participants pro- oncologic care, would have a better quality of life, vided written informed consent. The protocol, lower rates of depressive symptoms, and less including the statistical analysis plan, is avail- aggressive end-of-life care.
able at NEJM.org. All the authors attest that the study was performed in accordance with the protocol and the statistical analysis plan.
Study Design
From June 7, 2006, to July 15, 2009, we enrolled Patients who presented to the outpatient thoracic ambulatory patients with newly diagnosed meta- oncology clinic were invited by their medical on- n engl j med 363;8 nejm.org august 19, 2010 The New England Journal of Medicine Downloaded from www.nejm.org on September 2, 2010. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. Early Palliative Care for Metastatic Cancer cologists to enroll in the study; all the medical had to be present for more than half the time, oncologists in the clinic agreed to approach, re- except for the symptom of suicidal thoughts, cruit, and obtain consent from their patients. which was included in the diagnosis if it was Physicians were encouraged, but not required, to present at any time.
offer participation to all eligible patients; no ad- ditional screening or recruitment measures were Measures of Health Care Use
used. Patients were eligible to participate if they Data were collected from the electronic medical had pathologically confirmed metastatic non– record on the use of health services and end-of- small-cell lung cancer diagnosed within the pre- life care, including anticancer therapy, medication vious 8 weeks and an Eastern Cooperative Oncol- prescriptions, referral to hospice, hospital admis- ogy Group (ECOG) performance status of 0, 1, or sions, emergency department visits, and the date 2 (with 0 indicating that the patient is asymp- and location of death. Patients were classified as tomatic, 1 that the patient is symptomatic but having received aggressive care if they met any of fully ambulatory, and 2 that the patient is symp- the following three criteria: chemotherapy within tomatic and in bed <50% of the day)16 and were 14 days before death, no hospice care, or admis- able to read and respond to questions in English. sion to hospice 3 days or less before death.20-22 Patients who were already receiving care from Finally, we assessed whether patients' resuscita- the palliative care service were not eligible for tion preferences were documented in the outpa- participation in the study.
tient electronic medical record.23 Health-related quality of life was measured with Participants completed baseline questionnaires the use of the Functional Assessment of Cancer before randomization. Follow-up assessments of Therapy–Lung (FACT-L) scale, which assesses quality of life and mood were performed at 12 multiple dimensions of the quality of life (physi- weeks (or at an outpatient clinic visit within cal, functional, emotional, and social well-being) 3 weeks before or after that time point). Partici- during the previous week.17 In addition, the lung- pants who had no scheduled clinic visits within cancer subscale (LCS) of the FACT-L scale evalu- this period received the questionnaires by mail. ates seven symptoms specific to lung cancer. The When responses on questionnaires were incom- primary outcome of the study was the change plete, research staff documented the reasons for from baseline to 12 weeks in the score on the which the participant did not give a full response.
Trial Outcome Index (TOI), which is the sum of the scores on the LCS and the physical well-being Statistical Analysis
and functional well-being subscales of the FACT-L Data obtained through December 1, 2009, were included in the analyses. The primary outcome Mood was assessed with the use of both the was the change in the score on the TOI from Hospital Anxiety and Depression Scale (HADS) baseline to 12 weeks. We estimated that with 120 and the Patient Health Questionnaire 9 (PHQ- patients, the study would have 80% power to de- 9).18,19 The 14-item HADS, which consists of two tect a significant between-group difference in the subscales, screens for symptoms of anxiety and change in the TOI score from baseline to 12 depression in the previous week. Subscale scores weeks, with a medium effect size of 0.5 SD.24 The range from 0, indicating no distress, to 21, in- protocol was amended in August 2008 to allow dicating maximum distress; a score higher than for the enrollment of an additional 30 participants 7 on either HADS subscale is considered to be in order to compensate for the loss of any patients clinically significant. The PHQ-9 is a nine-item to follow-up.
measure that evaluates symptoms of major de- Statistical analyses were performed with the pressive disorder according to the criteria of the use of SPSS software, version 16.0 (SPSS). Descrip- fourth edition of the Diagnostic and Statistical Manual tive statistics were used to estimate the frequen- of Mental Disorders (DSM-IV). A major depressive cies, means, and standard deviations of the study syndrome was diagnosed if a patient reported at variables. Differences between study groups in least five of the nine symptoms of depression on baseline characteristics and clinical outcomes the PHQ-9, with one of the five symptoms being were assessed with the use of two-sided Fisher's either anhedonia or depressed mood. Symptoms exact tests and chi-square tests for categorical n engl j med 363;8 nejm.org august 19, 2010 The New England Journal of Medicine Downloaded from www.nejm.org on September 2, 2010. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. variables and independent-samples Student's t-tests 2009) were censored on that date. A Cox propor- for continuous variables. Multivariate linear re- tional-hazards model was used to assess the ef- gression analyses, adjusted for baseline scores, fect of early palliative care on survival, with ad- were used to examine the effect of early palliative justment for demographic characteristics and care on quality-of-life outcomes. For intention- baseline ECOG performance status.
to-treat analyses, we used the conservative method of carrying baseline values forward to account for all missing patient-reported outcome data, including data that were missing owing to death. Baseline Characteristics of the Patients
Survival time was calculated from the date of A total of 151 patients were enrolled in the study enrollment to the date of death with the use of (see the figure in the Supplementary Appendix). the Kaplan–Meier method. Data from patients who The percentage of patients enrolled was similar were alive at the last follow-up (December 1, for each of the thoracic oncologists in the clinic. Table 1. Baseline Characteristics of the Study Participants.*
Early Palliative Care
P Value†
Female sex — no. (%) Race — no. (%)‡ Hispanic or Latino ethnic group‡ Marital status — no. (%) Divorced or separated ECOG performance status — no. (%)¶ Presence of brain metastases — no. (%) Initial anticancer therapy — no. (%) Platinum-based combination chemotherapy Oral EGFR tyrosine kinase inhibitor Receipt of initial chemotherapy as part of a clinical trial — no. (%) Never smoked or smoked ≤10 packs/yr — no./ Assessment of mood symptoms — no./total no. (%) Depression subscale PHQ-9 major depressive syndrome†† n engl j med 363;8 nejm.org august 19, 2010 The New England Journal of Medicine Downloaded from www.nejm.org on September 2, 2010. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. Early Palliative Care for Metastatic Cancer Table 1. (Continued.)
Early Palliative Care
P Value†
Scores on quality-of-life measures‡‡ Lung-cancer subscale Trial Outcome Index * Plus–minus values are means ±SD. Percentages may not total 100 because of rounding. ECOG denotes Eastern Cooperative Oncology Group, EFGR epidermal growth factor receptor, FACT-L Functional Assessment of Cancer Therapy–Lung, HADS Hospital Anxiety and Depression Scale, and PHQ-9 Patient Health Questionnaire 9.
† P values were calculated with the use of two-sided chi-square and Fisher's exact tests for categorical variables and the independent-samples Student's t-tests for continuous variables.
‡ Race or ethnic group was self-reported.
§ The P value is for the between-group comparison of the proportions of patients who were white and those who were members of a minority group (black and Asian), calculated with the use of Fisher's exact test.
¶ An ECOG performance status of 0 indicates that the patient is asymptomatic, 1 that the patient is symptomatic but fully ambulatory, and 2 that the patient is symptomatic and in bed less than 50% of the day.
‖ The P value is for the between-group comparison of the proportion of patients receiving platinum-based combination chemotherapy and the proportion receiving other treatments, calculated with the use of Fisher's exact test.
** The HADS consists of two subscales, one for symptoms of anxiety and one for symptoms of depression. Subscale scores range from 0, indicating no distress, to 21, indicating maximum distress; a score higher than 7 indicates clinical- ly meaningful anxiety or depression.
†† The PHQ-9 is a nine-item measure that evaluates symptoms of major depressive disorder according to the criteria of the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV). A major depressive syndrome was diagnosed if a patient reported at least five of the nine symptoms of depression on the PHQ-9, with one of the five symptoms being either anhedonia or depressed mood. Symptoms had to be present for more than half the time, except for the symptom of suicidal thoughts, which was included in the diagnosis if it was present at any time.
‡‡ The quality of life was assessed with the use of three measures: the FACT-L scale, on which scores range from 0 to 136, with higher scores indicating a better quality of life; the lung-cancer subscale of the FACT-L scale, on which scores range from 0 to 28, with higher scores indicating fewer symptoms; and the Trial Outcome Index, which is the sum of the scores on the lung-cancer, physical well-being, and functional well-being subscales of the FACT-L scale (scores range from 0 to 84, with higher scores indicating a better quality of life).
No significant differences in demographic char- palliative care service by the 12th week. The aver- acteristics or overall survival were seen between age number of visits in the palliative care group the study participants and eligible patients who was 4 (range, 0 to 8). Ten patients who received were not enrolled in the study. The baseline char- standard care (14%) had a palliative care consul- acteristics were well matched between the two tation in the first 12 weeks of the study, primar- study groups (Table 1). Known prognostic factors, ily to address the management of symptoms, with including age, sex, ECOG performance status, seven patients having one visit and three having presence or absence of brain metastases, smoking two visits.
status, and initial anticancer therapy, were also balanced between the study groups. Although Quality-of-Life and Mood Outcomes
genetic testing was not routinely performed, the A comparison of measures of quality of life at 12 proportions of patients with mutations in the weeks showed that the patients assigned to early epidermal growth factor gene (EGFR) were simi- palliative care had significantly higher scores lar between the study groups among the patients than did those assigned to standard care, for the who underwent testing (9% in the palliative care total FACT-L scale, the LCS, and the TOI, with group and 12% in the standard-treatment group, effect sizes in the medium range (Table 2). Pa- P = 0.76). No significant between-group differenc- tients in the palliative care group had a 2.3-point es were seen in baseline quality of life or mood increase in mean TOI score from baseline to 12 weeks, as compared with a 2.3-point decrease in the standard care group (P = 0.04) (Fig. 1). With the use of linear regression to control for base- All the patients assigned to early palliative care, line quality-of-life values, the group assignment except for one patient who died within 2 weeks significantly predicted scores at 12 weeks on the after enrollment, had at least one visit with the total FACT-L scale (adjusted difference in mean n engl j med 363;8 nejm.org august 19, 2010 The New England Journal of Medicine Downloaded from www.nejm.org on September 2, 2010. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. Table 2. Bivariate Analyses of Quality-of-Life Outcomes at 12 Weeks.*
Difference between Early
Early Palliative Care
Care and Standard Care
P Value†
* Plus–minus values are means ±SD. Quality of life was assessed with the use of three scales: the Functional Assessment of Cancer Therapy–Lung (FACT-L) scale, on which scores range from 0 to 136, with higher scores indicating better qual- ity of life; the lung-cancer subscale (LCS) of the FACT-L scale, on which scores range from 0 to 28, with higher scores indicating fewer symptoms; and the Trial Outcome Index (TOI), which is the sum of the scores on the LCS and the physical well-being and functional well-being subscales of the FACT-L scale (scores range from 0 to 84, with higher scores indicating better quality of life).
† The P value was calculated with the use of two-sided Student's t-tests for independent samples.
‡ The effect size was determined with the use of Cohen's d statistic, which is a measure of the difference between two means (in this case, the mean in the group assigned to early palliative care group minus the mean in the group assigned to standard care) divided by a standard deviation for the pooled data. According to the conventional classification, an effect size of 0.20 is small, 0.50 moderate, and 0.80 large.
[±SE] scores, 5.4±2.4; 95% confidence interval for the comparison of patients with symptoms [CI], 0.7 to 10.0; P = 0.03) and the TOI (adjusted of major depression on the PHQ-9).
difference in mean scores, 5.2±1.8; 95% CI, 1.6 to 8.9; P = 0.005), but not on the LCS (adjusted differ- End-of-Life Care
ence in mean scores, 1.0±0.6; 95% CI, –0.2 to 2.3; At the time of the analysis of end-of-life care, 105 P = 0.12). In addition, the percentage of patients participants (70%) had died; the median duration with depression at 12 weeks, as measured by the of follow-up among participants who died was HADS and PHQ-9, was significantly lower in the 5.7 months. Within this subsample, a greater per- palliative care group than in the standard care centage of patients in the group assigned to stan- group, although the proportions of patients re- dard care than in the group assigned to early ceiving new prescriptions for antidepressant drugs palliative care received aggressive end-of-life care were similar in the two groups (approximately (54% [30 of 56 patients] vs. 33% [16 of 49 pa- 18% in both groups, P = 1.00) (Fig. 2). The per- tients], P = 0.05). In addition, fewer patients in the centage of patients with elevated scores for symp- standard care group than in the palliative care toms of anxiety did not differ significantly be- group had resuscitation preferences documented tween the groups.
in the outpatient electronic medical record (28% The figure in the Supplementary Appendix in- [11 of 39 patients who had preferences docu- cludes an explanation of missing data according mented during the course of the study] vs. 53% to study group. There was no significant asso- [18 of 34 patients], P = 0.05). The study did not ciation between missing data on patient-reported have adequate power to examine specific indica- outcomes at 12 weeks and any baseline charac- tors of aggressive care at the end of life. Howev- teristic (although there was a trend toward a er, analyses of various measures of utilization, significant association between missing data and such as rates of hospitalization and emergency assigned treatment [P = 0.07]). When we carried department visits (Table 2 in the Supplementary the baseline scores of the participants forward Appendix), as well as the duration of hospice care for the missing data on patient-reported out- (median duration, 11 days in the palliative care comes, all primary treatment effects were repli- group vs. 4 days in the standard care group; cated with respect to quality of life (P = 0.04 for P = 0.09 with the use of the Wilcoxon rank-sum the 12-week FACT-L score, P = 0.01 for the 12-week test), suggested an improvement in the quality of LCS score, P = 0.04 for the 12-week TOI score, care with early palliative care. Despite receiving and P = 0.04 for the mean change from baseline less aggressive end-of-life care, patients in the pal- to 12 weeks in the TOI score) and mood (P = 0.04 liative care group had significantly longer survival for the comparison of patients with elevated scores than those in the standard care group (median on the HADS depression subscale, and P = 0.02 survival, 11.6 vs. 8.9 months; P = 0.02) (Fig. 3).
n engl j med 363;8 nejm.org august 19, 2010 The New England Journal of Medicine Downloaded from www.nejm.org on September 2, 2010. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. Early Palliative Care for Metastatic Cancer Figure 1. Mean Change in Quality-of-Life Scores
from Baseline to 12 Weeks in the Two Study Groups.
Quality of life was assessed with the use of the Func-tional Assessment of Cancer Therapy–Lung (FACT-L) scale, on which scores range from 0 to 136, with higher scores indicating a better quality of life; the lung-cancer subscale (LCS) of the FACT-L scale, on which scores range from 0 to 28, with higher scores indicating fewer symptoms; and the Trial Outcome Index (TOI), which is the sum of the scores on the LCS and the physical Mean Change
well-being and functional well-being subscales of the FACT-L scale (scores range from 0 to 84, with higher scores indicating a better quality of life). With study group as the independent variable, two-sided indepen- dent-samples Student's t-tests showed a trend toward a significant between-group difference in the mean (±SD) change in scores from baseline to week 12 on the FACT-L scale (−0.4±13.8 in the standard care group vs. 4.2±13.8 in the palliative care group; difference be-tween groups, 4.6; 95% confidence interval [CI], −0.8 to 9.9; P = 0.09) (Panel A), no significant between-group difference in the mean change in scores on the LCS (0.3±4.0 and 0.8±3.6 in the two groups, respective- ly; difference between groups, 0.5; 95% CI, −1.0 to 2.0; P = 0.50) (Panel B), and a significant between-group difference in the mean change in scores on the TOI Mean Change
(−2.3±11.4 vs. 2.3±11.2; difference between groups, 4.6; 95% CI, 0.2 to 8.9; P = 0.04) (Panel C). Data are from the 47 patients in the standard care group and the 60 patients in the palliative care group who completed the 12-week assessments. I bars indicate 95% confidence This study shows the effect of palliative care when it is provided throughout the continuum of care for advanced lung cancer. Early integration of palliative care with standard oncologic care in patients with metastatic non–small-cell lung can- Mean Change
cer resulted in survival that was prolonged by ap- proximately 2 months and clinically meaningful improvements in quality of life and mood. More- over, this care model resulted in greater docu- mentation of resuscitation preferences in the outpatient electronic medical record, as well as less aggressive care at the end of life. Less ag- count for the observed survival benefit. In addi- gressive end-of-life care did not adversely affect tion, the integration of palliative care with stan- survival. Rather, patients receiving early pallia- dard oncologic care may facilitate the optimal tive care, as compared with those receiving stan- and appropriate administration of anticancer dard care alone, had improved survival. Previous therapy, especially during the final months of data have shown that a lower quality of life and life. With earlier referral to a hospice program, depressed mood are associated with shorter sur- patients may receive care that results in better vival among patients with metastatic non–small- management of symptoms, leading to stabiliza- cell lung cancer.25-27 We hypothesize that im- tion of their condition and prolonged survival. provements in both of these outcomes among These hypotheses require further study.
patients assigned to early palliative care may ac- Improving quality of life and mood in patients n engl j med 363;8 nejm.org august 19, 2010 The New England Journal of Medicine Downloaded from www.nejm.org on September 2, 2010. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. tion in the quality of life over time, which is Early palliative care consistent with the results in the standard care group in our study.30-32 Despite similar cancer therapies in our two study groups, the patients assigned to early palliative care had an improved quality of life, as compared with those receiving standard care. Rates of depression also differed significantly between the groups, with approxi- mately half as many patients in the palliative care group as in the standard care group reporting clinically significant depressive symptoms on the Patients with Mood Symptoms (%)
HADS, and this effect was not due to a between- group difference in the use of antidepressant Figure 2. Twelve-Week Outcomes of Assessments
To date, evidence supporting a benefit of pal- liative care is sparse, with most studies having Depressive symptoms were assessed with the use of the Hospital Anxiety and Depression Scale (HADS), notable methodologic weaknesses, especially with which consists of two subscales, one for symptoms respect to quality-of-life outcomes.8 One study of anxiety (HADS-A) and one for symptoms of depres- with sufficient power to examine quality-of-life sion (HADS-D) (subscale scores range from 0, indi- outcomes showed that among patients receiving cating no distress, to 21, indicating maximum distress; radiation therapy, a multidisciplinary intervention a score higher than 7 on either HADS subscale is con-sidered to be clinically significant) and with the use of focused on education, behavioral modification, the Patient Health Questionnaire 9 (PHQ-9). The PHQ-9 and coping style resulted in improvements in the is a nine-item measure that evaluates symptoms of quality of life.33 A recent study showed that Proj- major depressive disorder according to the criteria ect ENABLE (Educate, Nurture, Advise, Before Life of the fourth edition of the Diagnostic and Statistical Ends), a telephone-based, psychoeducational pro- Manual of Mental Disorders (DSM-IV). A major depres-sive syndrome was diagnosed if a patient reported at gram for patients with advanced cancer, signifi- least five of the nine symptoms of depression on the cantly improved both quality of life and mood.34 PHQ-9, with one of the five symptoms being either However, the percentage of patients who com- anhedonia or depressed mood. Symptoms had to be pleted the study assessments was somewhat low, present for more than half the time, except for the and the study did not use a traditional palliative symptom of suicidal thoughts, which was included in the diagnosis if it was present at any time. The percent- ages of patients with mood symptoms, assessed on the Our study also showed that early outpatient basis of each of these measures, in the group assigned palliative care for patients with advanced cancer to standard treatment and the group assigned to early can alter the use of health care services, including palliative care, respectively, are as follows: HADS-D, care at the end of life. Other studies of outpa- 38% (18 of 47 patients) versus 16% (9 of 57), P = 0.01; HADS-A, 30% (14 of 47 patients) and 25% (14 of 57), tient palliative care have failed either to investi- respectively; P = 0.66; and PHQ-9, 17% (8 of 47 patients) gate these outcomes or to show an effect on the versus 4% (2 of 57); P = 0.04. The analyses were per- use of resources.5,34,35 In our trial, significantly formed with the use of a two-sided Fisher's exact test.
more patients in the group assigned to early pal- liative care than in the standard care group had with metastatic non–small-cell lung cancer is a resuscitation preferences documented in the out- formidable challenge, given the progressive na- patient electronic medical record, an essential ture of the illness.28 The improvement we ob- step in clarifying and ensuring respect for pa- served in the quality of life among patients as- tients' wishes about their care at the end of signed to early palliative care, as indicated by a life.36 Early introduction of palliative care also mean change in the TOI score by 12 weeks that led to less aggressive end-of-life care, including was approximately 5 points higher in the pallia- reduced chemotherapy and longer hospice care. tive care group than in the standard care group, Given the trends toward aggressive and costly is similar to the improvement in the quality of care near the end of life among patients with life that has been observed among patients who cancer, timely introduction of palliative care may have a response to cisplatin-based chemothera- serve to mitigate unnecessary and burdensome py.29 Most studies show that there is a deteriora- personal and societal costs.20,37 n engl j med 363;8 nejm.org august 19, 2010 The New England Journal of Medicine Downloaded from www.nejm.org on September 2, 2010. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. Early Palliative Care for Metastatic Cancer Our study has several advantages over previ- ous studies, in which investigators have often re- lied on referrals to palliative care instead of us- ing a recruitment approach designed to obtain a representative sample.5,35 Because all patients with a new diagnosis of metastatic non–small-cell lung cancer were eligible for enrollment in our study, we extended the generalizability of our findings. Another strength of our trial was the low rate of loss to follow-up and the high per- centage of participants who completed the study assessments. In addition, the dropout rate by week 12 was less than 1%, further supporting Patients Surviving (%)
Early palliative care the feasibility and acceptability of early palliative care. Finally, the trial was adequately powered to detect changes in both quality of life and mood, and we prospectively collected data on end-of- Several limitations of the study deserve men- tion. It was performed at a single, tertiary care site with a specialized group of thoracic oncol- ogy providers and palliative care clinicians, there- Figure 3. Kaplan–Meier Estimates of Survival According to Study Group.
by limiting generalization of the results to other Survival was calculated from the time of enrollment to the time of death, care settings or patients with other types of if it occurred during the study period, or to the time of censoring of data on December 1, 2009. Median estimates of survival were as follows: 9.8 months cancer. In addition, because the sample lacked (95% confidence interval [CI], 7.9 to 11.7) in the entire sample (151 patients), diversity with respect to race and ethnic group, 11.6 months (95% CI, 6.4 to 16.9) in the group assigned to early palliative we were unable to assess the effect of these care (77 patients), and 8.9 months (95% CI, 6.3 to 11.4) in the standard important factors on study outcomes. Although care group (74 patients) (P = 0.02 with the use of the log-rank test). After we used a randomized, controlled design, both adjustment for age, sex, and baseline Eastern Cooperative Oncology Group performance status, the group assignment remained a significant predictor the patients and the clinicians were aware of the of survival (hazard ratio for death in the standard care group, 1.70; 95% CI, study assignments. To account for possible in- 1.14 to 2.54; P = 0.01). Tick marks indicate censoring of data.
fluences of care that are not specific to the pal- liative care provided, follow-up investigations the study participants who received standard should include a control group that receives a care, those who were assigned to early palliative similar amount of attention. In addition, we did care had improved mood, more frequent docu- not deny palliative care consultations to partici- mentation of resuscitation preferences, and less pants receiving standard care, and a small minor- aggressive end-of-life care. Although our find- ity of patients in the standard care group was ings must be replicated in a variety of care set- seen by the palliative care team. The data from tings and cancer populations, the results none- these patients were analyzed with the data from theless offer great promise for alleviating distress their assigned study group (standard care), a fac- in patients with metastatic disease and address- tor that may have diluted our findings. Finally, ing critical concerns regarding the use of health carrying the last observation forward for all miss- care services at the end of life.
ing data in the intention-to-treat analyses is a Supported by an American Society of Clinical Oncology Career conservative approach; therefore, the actual treat- Development Award and philanthropic gifts from the Joanne ment effect of early palliative care may be greater Hill Monahan Cancer Fund and Golf Fights Cancer.
Dr. Temel reports receiving payment for developing continu- than we report.
ing medical education (CME) programs from InforMEDical; and Early integration of palliative care for patients Dr. Lynch, serving on the board of Infinity Pharmaceuticals, re- with metastatic non–small-cell lung cancer is a ceiving consulting fees from Roche, Boehringer Ingelheim, Merck, AstraZeneca, Bristol-Myers Squibb, and Sanofi-Aventis, clinically meaningful and feasible care model royalties from Partners HealthCare, and payment for developing that has effects on survival and quality of life CME programs from InforMEDical. No other potential conflict that are similar to the effects of first-line chemo- of interest relevant to this article was reported. Disclosure forms provided by the authors are available with therapy in such patients.28,38,39 As compared with the full text of this article at NEJM.org.
n engl j med 363;8 nejm.org august 19, 2010 The New England Journal of Medicine Downloaded from www.nejm.org on September 2, 2010. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. Early Palliative Care for Metastatic Cancer References
1. Mongan JJ, Ferris TG, Lee TH. Options tients with incurable NSCLC. J Clin Oncol ive care setting of the Big Lung Trial in
for slowing the growth of health care 2009;27:Suppl:20635. abstract.
non-small-cell lung cancer. J Clin Oncol costs. N Engl J Med 2008;358:1509-14.
16. Oken MM, Creech RH, Tormey DC, et 2005;23:7417-27.
2. Weinstein MC, Skinner JA. Compara-
al. Toxicity and response criteria of the 29. Cella D, Eton DT, Fairclough DL, et al.
tive effectiveness and health care spend- Eastern Cooperative Oncology Group. Am What is a clinically meaningful change ing — implications for reform. N Engl J J Clin Oncol 1982;5:649-55.
on the Functional Assessment of Cancer Med 2010;362:460-5.
17. Cella DF, Bonomi AE, Lloyd SR, Tulsky Therapy-Lung (FACT-L) Questionnaire? Re-
3. Ferris FD, Bruera E, Cherny N, et al. DS, Kaplan E, Bonomi P. Reliability and sults from Eastern Cooperative Oncology
Palliative cancer care a decade later: validity of the Functional Assessment of Group (ECOG) Study 5592. J Clin Epide- accomplishments, the need, next steps — Cancer Therapy-Lung (FACT-L) quality of miol 2002;55:285-95.
from the American Society of Clinical On- life instrument. Lung Cancer 1995;12:199- 30. Bonomi P, Kim K, Fairclough D, et al.
cology. J Clin Oncol 2009;27:3052-8.
Comparison of survival and quality of life 4. Levy MH, Back A, Benedetti C, et al. 18. Zigmond AS, Snaith RP. The Hospital in advanced non-small-cell lung cancer
NCCN clinical practice guidelines in on- Anxiety and Depression Scale. Acta Psy- patients treated with two dose levels of cology: palliative care. J Natl Compr Canc chiatr Scand 1983;67:361-70.
paclitaxel combined with cisplatin versus Netw 2009;7:436-73.
19. Kroenke K, Spitzer RL, Williams JB. etoposide with cisplatin: results of an
5. Follwell M, Burman D, Le LW, et al. The PHQ-9: validity of a brief depression Eastern Cooperative Oncology Group trial.
Phase II study of an outpatient palliative severity measure. J Gen Intern Med 2001; J Clin Oncol 2000;18:623-31.
care intervention in patients with meta- 16:606-13.
31. Langer CJ, Manola J, Bernardo P, et al.
static cancer. J Clin Oncol 2009;27:206- 20. Earle CC, Landrum MB, Souza JM, Cisplatin-based therapy for elderly patients
Neville BA, Weeks JC, Ayanian JZ. Aggres- with advanced non-small-cell lung cancer: 6. Jordhøy MS, Fayers P, Loge JH, Ahlner-
siveness of cancer care near the end of implications of Eastern Cooperative On- Elmqvist M, Kaasa S. Quality of life in life: is it a quality-of-care issue? J Clin On- cology Group 5592, a randomized trial. palliative cancer care: results from a clus- col 2008;26:3860-6.
J Natl Cancer Inst 2002;94:173-81.
ter randomized trial. J Clin Oncol 2001; 21. American Society of Clinical Oncol- 32. Sandler AB, Nemunaitis J, Denham C,
ogy Quality Oncology Practice Initiative. et al. Phase III trial of gemcitabine plus 7. Morita T, Akechi T, Ikenaga M, et al. ASCO QOPI summary of measures. Fall cisplatin versus cisplatin alone in patients
Late referrals to specialized palliative care 2009. (Accessed July 23, 2010, at http:// with locally advanced or metastatic non- service in Japan. J Clin Oncol 2005;23: qopi.asco.org/Documents/QOPIFall2009 small-cell lung cancer. J Clin Oncol 2000; 8. Zimmermann C, Riechelmann R, 22. Tang ST, Wu SC, Hung YN, Chen JS, 33. Rummans TA, Clark MM, Sloan JA, et
Krzyzanowska M, Rodin G, Tannock I. Huang EW, Liu TW. Determinants of ag- al. Impacting quality of life for patients Effectiveness of specialized palliative care: gressive end-of-life care for Taiwanese with advanced cancer with a structured a systematic review. JAMA 2008;299: cancer decedents, 2001 to 2006. J Clin multidisciplinary intervention: a random- ized controlled trial. J Clin Oncol 2006; 9. Jemal A, Siegel R, Ward E, Hao Y, Xu J, 23. Temel JS, Greer JA, Admane S, et al. 24:635-42.
Thun MJ. Cancer statistics, 2009. CA Can- Code status documentation in the outpa- 34. Bakitas M, Lyons KD, Hegel MT, et al.
cer J Clin 2009;59:225-49.
tient electronic medical records of pa- Effects of a palliative care intervention 10. Hopwood P, Stephens RJ. Depression tients with metastatic cancer. J Gen In-
on clinical outcomes in patients with ad- in patients with lung cancer: prevalence tern Med 2010;25:150-3.
vanced cancer: the Project ENABLE II ran- and risk factors derived from quality-of- 24. Cella D. The Functional Assessment of domized controlled trial. JAMA 2009;302:
life data. J Clin Oncol 2000;18:893-903.
Cancer Therapy–Lung and Lung Cancer 741-9.
11. Hopwood P, Stephens RJ. Symptoms Subscale assess quality of life and mean-
35. Rabow MW, Dibble SL, Pantilat SZ,
at presentation for treatment in patients ingful symptom improvement in lung can- McPhee SJ. The comprehensive care team: with lung cancer: implications for the cer. Semin Oncol 2004;31:Suppl 9:11-5.
a controlled trial of outpatient palliative evaluation of palliative treatment. Br J 25. Maione P, Perrone F, Gallo C, et al. medicine consultation. Arch Intern Med
Pretreatment quality of life and function- 2004;164:83-91.
12. Lutz S, Norrell R, Bertucio C, et al. al status assessment significantly predict 36. Walling A, Lorenz KA, Dy SM, et al.
Symptom frequency and severity in pa- survival of elderly patients with advanced Evidence-based recommendations for in- tients with metastatic or locally recurrent non-small-cell lung cancer receiving che- formation and care planning in cancer lung cancer: a prospective study using the motherapy: a prognostic analysis of the care. J Clin Oncol 2008;26:3896-902.
Lung Cancer Symptom Scale in a com- Multicenter Italian Lung Cancer in the 37. Emanuel EJ, Ash A, Yu W, et al. Man-
munity hospital. J Palliat Med 2001;4:157- Elderly Study. J Clin Oncol 2005;23:6865- aged care, hospice use, site of death, and medical expenditures in the last year of 13. Temel JS, Jackson VA, Billings JA, et al. 26. Movsas B, Moughan J, Sarna L, et al. life. Arch Intern Med 2002;162:1722-8.
Phase II study: integrated palliative care Quality of life supersedes the classic prog- 38. Non-Small Cell Lung Cancer Collab-
in newly diagnosed advanced non-small- nosticators for long-term survival in local- orative Group. Chemotherapy in non-small cell lung cancer patients. J Clin Oncol ly advanced non-small-cell lung cancer: cell lung cancer: a meta-analysis using an analysis of RTOG 9801. J Clin Oncol updated data on individual patients from 14. Clinical practice guidelines for quality 2009;27:5816-22.
52 randomised clinical trials. BMJ 1995; palliative care. 2nd ed. Pittsburgh: Nation- 27. Pirl WF, Temel JS, Billings A, et al. 311:899-909.
al Consensus Project, 2009. (Accessed July Depression after diagnosis of advanced 39. Spiro SG, Rudd RM, Souhami RL, et
23, 2010, at www.nationalconsensusproject non-small cell lung cancer and survival: al. Chemotherapy versus supportive care a pilot study. Psychosomatics 2008;49: in advanced non-small cell lung cancer: 15. Jackson VA, Jacobsen JC, Greer J, et al. 218-24.
improved survival without detriment to Components of early intervention out- 28. Brown J, Thorpe H, Napp V, et al. As- quality of life. Thorax 2004;59:828-36.
patient palliative care consultation in pa- sessment of quality of life in the support- Copyright 2010 Massachusetts Medical Society. n engl j med 363;8 nejm.org august 19, 2010 The New England Journal of Medicine Downloaded from www.nejm.org on September 2, 2010. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved.

Source: http://www.cuentaconnosotros.es/wp-content/uploads/2010/09/Early-PC-NEJM.pdf