Early palliative care improves quality of life, reduces aggressiveness of care at the end-

of-life and prolongs survival in stage IV NSCLC patients:

Results of a phase III randomized trial

Temel JS, Greer JA, Gallagher E, Admane S, Pirl WF, Jackson VA, Dahlin C, Muzikansky A, Jacobsen J, Lynch

TJ

Study Rationale

Current Care Model

Proposed Care Model

www.iom.edu

50 patients with newly diagnosed metastatic NSCLC

Meet with palliative care at least monthly during 1st 6 months of

care

Temel, JCO (25)17, 2007

Integrated Oncology and Palliative Care in the Ambulatory Care Setting

150 patients

with newly diagnosed metastatic

NSCLC

Early palliative care integrated with standard oncology care

Standard oncology care

Baseline Data Collection

RANDOMIZED

Study Design

Meet with palliative care within 3 weeks of signing consent and at least monthly thereafter

Meet with palliative care only when requested by patient, family or oncology clinician.

Early Palliative Care Study Procedures

Palliative Care Guidelines

Illness understanding and education Inquire about illness and prognostic understanding Offer clarification regarding treatment goals

Symptom management Pain Pulmonary symptoms Fatigue and sleep disturbance Mood Gastrointestinal

Decision-making Assess mode of decision-making Assist with treatment decision-making

Coping with life-threatening illness Patient Family/family caregivers

www.nationalconsensusproject.org

Study Objectives

Primary Objective: Measure the difference in QOL between the

two study arms at 12 weeks.

Secondary Objectives: 1. Psychological distress at 12 weeks2. Quality of end-of-life care 3. Resource utilization at the end-of-life4. Documentation of resuscitation preference in

the medical record

Study Eligibility

1. Metastatic NSCLC diagnosed within the previous 8 weeks.

2. ECOG performance status 0-2.3. Ability to read and respond to

questions in English.4. Planning to receive oncology care at

the participating institution.

Study Measures

Quality of life FACT-Lung - Lower scores indicative of greater symptom

burden Lung Cancer Symptom (LCS): lung cancer specific symptoms Trial Outcome Index (TOI): LCS and functional and physical well-

being

Psychological Distress Hospital Anxiety and Depression Scale (HADS)

Score of > 8 of each subscale indicative of symptoms of depression or anxiety

Patient Health Questionnaire-9 (PHQ-9) Evaluates symptoms of major depressive disorder (MDD) using

DSM-IV criteria.

Data Collection

Measures of health care utilization were collected from electronic medical records. Use of anti-cancer therapies Hospital and emergency room visits Dates of hospice referral Date and location of death Documentation of resuscitation preference

Sample Size Calculation

Sample size was calculated to detect a clinically meaningful change in QOL, defined as a medium effect size of 0.5 SD.

120 patients were required to have 80% power to detect an effect size of 0.5 SD in FACT-Lung TOI.

Due to rapid accrual, the study was amended to add an additional 30 patients.

Data were analyzed through 12/1/09.

Statistical Analysis

Differences between study arms in clinical outcomes were assessed with two-sided Fisher’s Exact tests for categorical variables and independent-samples t-tests for continuous variables.

For ITT analyses, baseline values were carried forward for missing patient-reported outcome data.

Survival time was calculated from the date of consent to date of death using the Kaplan-Meier method.

Differences in survival were tested with Log Rank and Cox Proportional Hazard Model.

Study Flow

Assessed for eligibility (N=283)

June 2006 – July 2009

Excluded (n=9)Not offered (n=60)Refused to participate (n=59)Study closed during eligibility (n=4)

Randomly assigned (N=151)

Palliative care (N=77) Standard care (N=74)

12 week QOL assessment:47 completed (64%)17 died (23%)10 did not complete (13%)

12 week QOL assessment:60 completed (78%)10 died (13%)7 not completed (9%)

Sample Demographics Standard Care N=74

Mean (SD) or N (%)

Early Palliative Care N=77

Mean (SD) or N (%)

p-value

Age Mean (SD) 64.9 (9.4) 65.0 (9.7) 0.94

Female 36 (49) 42 (55) 0.47

Race White African American Asian

70 (95)3 (4)1 (1)

77 (100)0 (0)0 (0)

0.06

Ethnicity Hispanic 1 (1) 1 (1) 1.00

Marital Status Married Single Divorced/Separated Widowed

45 (61)9 (12)12 (16)8 (10)

48 (62)9 (12)12 (16)8 (10)

1.00

Patient Demographics

Clinical Characteristics Standard Care N=74 N (%)

Early Palliative CareN=77N (%)

p-value

ECOG PS: 0 1 2

30 (41)35 (47)9 (12)

26 (34)46 (60)5 (6)

0.24

Brain Metastasis: Yes No

19 (26)55 (74)

24 (31)53 (69)

0.46

Initial Anticancer Therapy: Platinum-Based Regimen Single Agent Chemotherapy Oral EGFR-TKI Radiation Both Chemo and Radiation No Chemotherapy

35 (47)3 (4)6 (8)

26 (35)3 (4)1 (1)

35 (45)9 (12)6 (8)

27 (35)0 (0)0 (0)

0.82

Type of Initial Chemotherapy Regimen: Standard Therapy Clinical Trial

35 (75) 12 (25)

36 (72)14 (28)

0.78

Baseline Clinical Characteristics

Variable Standard CareM (SD) or N (%)

Early Palliative Care

M (SD) or N (%)

p-value

FACT-Lung 91.7 (16.7) 93.6 (16.5) 0.50

Trial Outcome Index (TOI) 55.3 (13.1) 56.2 (13.4) 0.67

HADS Symptoms

Depression 18 (25) 17 (22) 0.67

Anxiety 24 (33) 28 (36) 0.70

Major Depressive Disorder 12 (17) 9 (12) 0.40

Baseline Quality of Life and Psychological Distress

Palliative Care Visits by 12 Weeks

Palliative Care Visits Standard Care (N=74)N (%)

Early Palliative Care (N=77)N (%)

None 64 (87) 1 (1)*

1 7 (9) 0

2 3 (4) 8 (10)

3 0 18 (23)

4 0 26 (34)

> 5 0 24 (31)* Died within 2 weeks of enrollment

12-week Quality of Life Measures

Variable Standard CareN=47

M (SD)

Early Palliative CareN=60

M (SD)

Effect Size

(Cohen’s d)

p-value

FACT-Lung 91.5 (15.8) 98.0 (15.1) 0.42 .03

Lung Cancer Symptoms (LCS)

19.3 (4.2) 21.0 (3.9) 0.42 .04

Trial Outcome Index (TOI)

53.0 (11.5) 59.0 (11.6) 0.52 0.009

Effect of Early PC on 12-week Psychological Distress

0

10

20

30

40

50

60

Depression Major DepressiveDisorder

Anxiety

Standard Care

Early Palliative Care

p=0.01

p=0.66

p=0.04

Change in QOL from Baseline to 12 Weeks

Mean change Early Palliative Care = + 2.3

Mean change Standard Care = - 2.3

p=0.04

Mean change Early Palliative Care = + 4.2

Mean change Standard Care = - 0.4

p=0.09

FACT-Lung FACT- Lung TOI

Quality of EOL Care and Resource Utilization

ASCO Quality Measures 1. No hospice2. Enrolled in hospice < 3 days before death 3. Chemotherapy within 14 days of death (DOD)

Measure Standard CareN (%) or Median

Early Palliative Care

N (%) or Median

p-value

Aggressive EOL Care No hospice Hospice < 3 days Chemo within 14 DOD

30 (54)22 (39)5 (15) 12 (24)

16 (33)15 (31)1 (3) 7 (18)

0.05

Hospital/ER Admissions within 30 DOD

31 (55) 19 (39) 0.12

Days on hospice 4 (0-269) 11 (0-117) 0.09

Documented Resuscitation Preference

11 (28) 18 (53) 0.05105 deaths at time of data analysis with data on chemotherapy within 14 DOD available on 90 patients

Standard care

Survival Analysis

Months

Overa

ll su

rviv

al

Median SurvivalEarly palliative care 11.6 moStandard care 8.9 mo

p=0.02Early palliative care

Controlling for age, gender and PS, adjusted HR=0.59 (0.40-0.88), p=0.01

Study Limitations

Single, tertiary-care site with a specialized group of clinicians.

Study population lacked racial/ethnic diversity. Randomized design but no blinding to study

arm. Small number of patients on standard care

arm seen by palliative care team. Survival was not a pre-specified study

endpoint. Lack of information on mediators of patient-

reported and medical outcomes.

Summary

Compared with standard oncology care, integrated palliative care led to: Improvements in QOL Lower rates of depression Less aggressive care at the end-of-life Greater documentation of resuscitation

preferences Higher survival rates

Discussion

Changes in QOL may be due to improved symptom management.

Decreased rates of depression may be related to improved symptom management and illness acceptance.

Prolonged survival possibly related to: Earlier recognition and management of medical

issues Improved QOL and mood Less chemotherapy at the end-of-life Longer hospice admissions

Acknowledgements

Funding Provided by: ASCO Foundation Golf Fights Cancer The Joanne Hill Monahan Fund

Supportive Care Research Group at Massachusetts General Hospital

William Pirl, MD, MPH Joseph Greer, Ph.D Inga Lennes MD Emily Gallagher, BS Sonal Admane, MBBS, MPH Elyse Park, Ph.D Areej El-Jawahri, MD

Center for Palliative Care at Massachusetts General Hospital

Andrew J. Billings MD Vicki Jackson MD Connie Dahlin ANP Craig Blinderman, MD Juliet Jacobsen, MD Amelia Cullinan, MD Sandy Nasrallah, MD

Thoracic Oncology at Massachusetts General Hospital

Panos Fidias, MD Alice Shaw, MD, Ph.D Rebecca Heist, MD Lecia Sequist, MD Jeff Engelman, MD, Ph.D David Barbie, MD, Ph.D. Inga Lennes, MD Elizabeth Lamont, MD Jeanne Vaughn, ANP Diane Doyle, ANP Patricia Ostler R.N Thoracic Oncology research nurses,

administrators and staff

Yale Cancer Center Thomas J. Lynch, MD