Annual Meeting Proceedings Part 1 - American Society of Clinical ...

598s Patient and Survivor Care
9126 General Poster Session (Board #48H), Sat, 8:00 AM-12:00 PM
Using scenarios to explain survival time: Attitudes of people with a cancer
experience. Presenting Author: Martin R. Stockler, NHMRC Clinical Trials
Centre, University of Sydney, Sydney, Australia
Background: Most people with advanced cancer want some information on
their life expectancy. We sought the attitudes of people with a cancer
experience to using three scenarios for survival (best-case, worst-case, and
typical) to explain life expectancy to people with advanced cancer.
Methods: Oncology clinic attendees and Breast Cancer Network Australia
(BCNA) members completed a survey describing two formats for explaining
life expectancy to a hypothetical patient with incurable cancer - providing
either three scenarios for survival or just the median survival time.
Associations between respondent attitudes and their demographic and
tumour characteristics were explored. Results: Characteristics of the505
respondents from outpatient clinics (n�251) and BCNA (n�254) were:
median age 58 years; female 74%; completed high school education 80%;
breast primary 64%; self-report of cancer spread to other body parts 23%
and; a median of 4 years since diagnosis of cancer. More respondents
agreed that explaining three scenarios (vs. just the median survival time)
would: make sense (93% vs. 75%), be helpful (93% vs. 69%), convey hope
(68% vs. 44%), reassure (60% vs. 40%) and help family members (91%
vs. 71%); all p-values �0.001. Fewer respondents agreed that explaining
three scenarios (vs. median survival) would upset people (24% v 36%);
p�0.001. The proportions of respondents agreeing that each of the 3
scenarios should be presented were: best-case 89%, worst-case 82%, and
typical 92%. Most respondents (85%) agreed it would be helpful to receive
a printed summary of the scenarios for survival. For information about their
own prognosis, 88% preferred all three scenarios and 5% a single estimate
of the median. Respondents with higher education were more likely to agree
that presenting three scenarios would be helpful (95% v 90%, p�0.05).
Respondents with breast cancer were more likely to agree that explaining
three scenarios would upset people (31% v 13%, p�0.001). Conclusions:
Most respondents judged presentation of best-case, worst-case and typical
scenarios preferable and more helpful and reassuring than presentation of
just the median survival time when explaining life expectancy to patients
with advanced cancer.
9128 General Poster Session (Board #49B), Sat, 8:00 AM-12:00 PM
Caring for people at the end of life: How do cancer caregivers differ from
other caregivers? Presenting Author: Afaf Girgis, University of New South
Wales, Sydney, Australia
Background: The size of the population in receipt of cancer care is growing
as the population ages and grows, with many receiving outpatient care,
living for longer and wishing to be cared for at home. Hence, cancer is now
one of the most common health conditions in receipt of informal care
giving. Methods: The South Australian Health Omnibus is an annual,
face-to-face, cross-sectional, whole-of-population, multi-stage, systematic
area sampling survey which seeks a minimum of 3,000 respondents each
year state-wide. From 2000-2007, questions about end of life were asked.
This study compares people who cared for someone with cancer until death
(cancer caregivers) with caregivers of people with other life limiting
illnesses (non-cancer caregivers). Results: 4,665/14,624 (31.9%; participation
rate 71.6%) had someone close to them die from an “expected”
death in the 5 years prior to being surveyed. One in 10 (10%; 1,504/
14,624) provided hands-on (day-to-day or intermittent) care; the majority
(79.5%) having cared for someone with cancer. Compared to non-cancer
caregivers, cancer caregivers cared for someone who was significantly
younger (mean age 66 vs 74 years; p�0.0001), and to have had a palliative
care service involved in the care of that person (64.9% vs 39%; p�0.000).
Whilst not statistically significant, cancer caregivers were somewhat less
likely to be the spouse of the person they cared for (11.8% vs 16.8%); more
likely to be of non-English speaking background (11% vs 7.5%); more
likely to report that the deceased was comfortable in the last 2 weeks of life
(44.1% vs 31.7%); and prepared to care again for someone with a
life-limiting illness (81.3% vs 71.4%). Conclusions: Caring for someone
with cancer at the end of life appears to be fundamentally different to other
caregiver populations, particularly in relation to age of, and relationship to,
the patient, which may contribute to their substantially greater utilisation
of palliative care services. Being a younger caregiver increases the
likelihood of them caring again in the future.
9127 General Poster Session (Board #49A), Sat, 8:00 AM-12:00 PM
A prospective study of family conferences (FCs) in a palliative care unit
(PCU) at a comprehensive cancer center. Presenting Author: Rony Dev,
University of Texas M. D. Anderson Cancer Center, Houston, TX
Background: TheFC provides the family information and emotional support
in the transition to end-of-life care. There are a few, mainly qualitative,
studies of FCs in a palliative care setting. A retrospective report suggested
increased emotional expression by family members when patients were not
present in the FC. The purpose of our study was to determine the
characteristics of FCs and the effect of patient participation on emotional
expression by family members. Methods: A prospective study was conducted
during 18-month period with140 consecutive FCs conducted by an
interdisciplinary team (palliative care physician, social worker, and others).
Data collected included demographics, discharge disposition, number of
participants, caregiver characteristics, expressions of emotional distress,
conflict with healthcare providers, and topics discussed. Descriptive
statistics and frequencies were calculated. Analysis was performed using a
chi-square test. Results: 70 patients (50%) were female, 64 (46%) white,
127 (91%) with solid tumors and a median age of 59. In 68/140 family
meetings (49%), patients actively participated. The primary caregivers
were predominantly female (66%), white (49%), and either the spouse or
partner (59%). A median number of 4 family members and 4 healthcare
providers attended the FCs with median duration of 1 hour and median of 2
days prior to discharge. Questions concerning advanced directives, symptoms
anticipated at death and caregiver well-being were infrequent.
Patients verbalized distress frequently, (73.1%). Primary caregivers emotional
expression of verbal distress were high but not significantly affected
by patient presence (82% vs 82%, P:NS); however, verbal expressions of
emotional distress by family were more common when patients were absent
(87%) than when present (73%), p � 0.037. Conclusions: There was a high
frequency of expression of emotional distress by patients and family
members. Patient participation was associated with decreased verbal
emotional expression by family members but not the primary caregiver.
Further studies are needed on the benefits of allocating additional time to
meet with family members without patient presence during a FC.
9129 General Poster Session (Board #49C), Sat, 8:00 AM-12:00 PM
Polymorphisms in genes OPRM1, �-arrestin 2, STAT6, and COMT and
clinical outcome to opioid therapy. Presenting Author: Sara Cros, Catalan
Institute of Oncology, Hospital Germans Trias i Pujol, Barcelona, Spain
Background: Morphine sulfate is the most widely used opioid in oncologic
pain. There is a high variability in response between individuals. The
objectives are to correlate the variability in response (R) and tolerability
with polymorphisms within OPRM1, �-arrestin 2, STAT6 and COMT gene
sequences. Methods: DNA from 44 oncologic patients (pts) with an EVA
score � 6 was prospectively analyzed using allelic discrimination techniques
and automatic sequencing. Pts (never treated with opioids) received
morphine sulfate 10 mg per day. Pain was assessed by EVA score and
toxicity was evaluated according to the most common opioid side effects.
Symptomatology was reassessed after 72 hours. Chi-squared and Fisher’s
tests where used to analyze data. Results: 17 pts (38.6%) responded with
an EVA � 3. 6 pts (13.6%) did not experience toxicity, 36 pts (81.8%) had
mild side effects and 2 pts (4.5%) had severe side effects. There was no
relation between R rate (RR) and tolerance. 76.5% of pts responding to
treatment had a CC/CT �-arrestin 2 genotype and this was associated with a
better RR (CC/CT R � 46.4% vs. TT R � 25% p � 0.16). The 2 pts having
severe side effects belonged to CC/CT group. COMT genotypes AA/GA were
associated with a higher RR (83.3%pts) and better tolerance (100% pts
with no toxicities). OPRM1 genotype GG was associated with both treatment
failure (0% responded) and worse tolerance (any toxicity � 100%;
severe side effects � 33%). Only 2,4% pts in the AA/GA genotype
experienced some toxicity (p � 0.041). Although STAT6 genotypes were
not associated with RR, tolerance was better in those pts with an AA or GA
genotype with 97.5% pts showing either mild or no side effects (p �
0.007). Conclusions: Common variants in �-arrestin 2, COMT, OPRM1 and
STAT6 genes have shown to predict for opioid-associated response and side
effects in oncologic patients. Further studies are needed to assess factors
determining the variability in response and tolerance to opioid treatment.
Visit abstract.asco.org and search by abstract for the full list of abstract authors and their disclosure information.