Symptom Benefit Study Measuring the Benefit of Palliative Chemotherapy in women with platinum refractory/ resistant ovarian cancer

1
Symptom Benefit StudyMeasuring the Benefit of
Palliative Chemotherapy in women with platinum
refractory/ resistant ovarian cancer
ANZGOG 0701
2
Study Background

• The aim is to develop a method to measure the
  benefit of chemotherapy, which takes into account
  BOTH subjective and objective responses
• Document time to symptom progression as an
  additional endpoint as well as symptom benefit
• Better insight into patterns of care and reasons
  for treatment with platinum resistant or
  refractory ovarian cancer
• Develop a prognostic index that better defines
  outcomes and test in a separate group

3
Objective

• Stage 1 To determine the symptoms and aspects of
  HRQL that are rated most severe, troublesome in
  patients and identify best instruments to use in
  stage 2
• Stage 2 To determine the proportion of women
  benefiting from palliative chemotherapy as
  defined by a clinically significant improvement
  in HRQL scores and improvement of symptoms and
  time to symptom progression.

4
Makhija S et al. Proc ASCO 2007Abstract 5507
5
Study Schema

• During Trial
• Stage1
• Complete QoL
• questionnaires at
• each cycle
• 20 subjects will
• participate in
• additional QoL
• telephone interviews
• Stage2
• Determine the optimal
• QoL forms from Stage1
• Longer follow-up
• Prognostic data collected at baseline

Target Population gt18yrs platinum
resistant/ refractory epithelial ovarian
cancer ECOG 0-3 Able to commence treatment
within 2wks of registration Ability to complete
QoL forms independently
Data Collection 4 Treatment cycles or
Disease progression Proposed longer follow-up
for Stage 2
REGISTER
6
Lung Cancer as a Model
Close parallels between platinum resistant/
refractory ovarian cancer and recurrent NSCLC and
SCLC in terms of response rates and survival
7
Fig 1. Survival in weeks TOPOTECAN VS. CAV
von Pawel, J. et al. J Clin Oncol 17658 1999
8
Symptom improvement compared with baseline in
patients with SCLC treated with i.v. topotecan or
CAV
Symptom i.v. topotecan CAV p value

Anorexia 32 16 0.042
Chest pain 25 17 0.371
Cough 25 15 0.160
Dyspnea 28 7 0.002
Fatigue 23 9 0.032
Hemoptysis 27 33 0.706
Hoarseness 33 13 0.043
Insomnia 33 19 0.085
Interference with daily activities 27 11 0.023


9
Jassem et al. 2002 33 Osoba et al. 1985 36 Fernandez et al. 1989 38 Hardy et al. 1989 39 Ellis et al. 1995 34 Tummarello et al. 1995 37 Cullen et al. 1988 35

Regimen GP BEP PVM/I MVbP MVbP PMVb MIP
Overall response () 41 44 42 21 32 33 56
Overall symptom improvement ()a 67 69 54
Symptom improvement () Symptom improvement () Symptom improvement () Symptom improvement () Symptom improvement () Symptom improvement () Symptom improvement () Symptom improvement ()
    Anorexia 50 58
    Cough 44 68 45 71 66 40 70
    Dyspnea 36 31 78 65 59 66 46
    Hemoptysis 75 78 91 100 92
    Malaise 53 53 62
    Pain 68 47 63 60 39 77
    Weight loss 44 89 30

Non Small Cell Lung Cancer Studies Gralla
Oncologist 2004 914-24
10
LCSS in relation to standard efficacy measures-
Maximum Improvement from baseline LCSS items- 2nd
line therapy
De marinis et al JTO 31 2008
11
LCSC in relation to standard efficacy measures
De marinis et al JTO 31 2008
12
Lung Cancer Studies

• Confirm that Symptom Benefit is a valid and
  valuable endpoint
• Correlation of Symptom Benefit with Response and
  SD
• Instruments sensitive to detect symptom benefit
• Provides complimentary efficacy data

13
Current status

• Fourteen sites open to recruitment
• Twelve in Australia
• Two in Canada
• A further nine Australian sites are currently
  awaiting final ethics approval
• Total recruitment
• 46
• 26 Australia
• 20 Canada

14
Baseline Demographics
Reason for treatment at enrolment
Symptom control/palliation rising CA125 radiological progression 28
Rising CA125 radiological progression 9
Symptom control/palliation rising CA125 6
Symptom Control radiological evidence 1
Radiological Evidence only 1
Rising CA125 only 1
N 46
15
Baseline Demographics contd

• Major symptoms reported at baseline
• 1. Pain
• 2. Fatigue
• 3. Abdominal Bloating
• ECOG 0 17 (N 45 - missing data for
  one patient)
• 1 26
• 2 2
• 3 0

16
Previous lines of chemotherapy
1 x line 2 x lines 3 x lines 4 x lines 5 x lines 7 x lines
18 8 9 5 2 1

N 43 Missing data on 3 x pts awaiting
response from sites
17

• Majority Platinum Resistant
• Compliance
• All questionnaires were completed to a very high
  compliance rate with few or no missing data

18
Death/Disease Progression
Cycle 2 Cycle 3 Cycle 4
Deaths 4 1
Disease progression 1 6
Completed 34 22 14

• 1 x pt. has withdrawn consent
• 3 x pts. off study due to site error
• NB. Due to centralised data entry, there is a
  time lag in receipt of CRFs

19
Stage 1 QoL Questionnaires

• Symptom Representation Questionnaire
• FACT-O (includes FOSI)
• EORTC QLQ-C30
• EORTC QLQ-OV28
• Patient Data Form
• Expected and Perceived Benefit Scale
• HAD Scale (Baseline End of Treatment only)
• Herth Hope Index (Baseline End of Treatment
  only)

20
Ovarian Symptom Benefit Study (OSBS) Initial
results and rationale for choosing the FACT-O
and revising the Patient DATA Form Ovarian as
the patient reported outcome measures (PRO) for
stage 2

• Prepared by Madeleine King and Martin Stockler on
  behalf of theOvarian Symptom Benefit Study Team
• 6 October 2009

21
Background Whats the problem?

• OSBS stage 1 uses 4 questionnaires to measure
  symptoms and/or quality of life
• items
• Symptom Representation Questionnaire (SRQ) 66
• Pt Disease Treatment Assessment (Pt DATA Form)
  48
• FACT-O (including 8 items of FOSI) 39
• QLQ-C30 Ov28 58
• The booklet was deliberately long and repetitive
  to corroborate findings and help determine the
  best subset of items for future studies
• Interviews with 10 patients indicated that they
  neither preferred nor disliked any particular
  questionnaires

22
Substitutability of candidate questionnaires

• SRQ vs. Patient DATA Form
• designed to measure clinically important aspects
  of QOL
• similar layouts, 0 to 10 scales
• single-item scoring (rather than multi-item or
  domain scoring)
• QLQ-C30/Ov28 vs. FACT-O
• designed to measure quality of life (QOL) in
  cancer clinical trials
• similar format and layout, similar response
  scales
• multi-item domains scoring (QLQ-C30 incl. some
  single items)
• We decided to choose for OSBS stage 2
• Either SRQ or Pt-DATA Form for individual
  symptoms
• Either QLQ-C30/Ov28 or FACT-O for
  multi-dimensional QOL

23
First filter analysis aims methods

• Aim Determine which 4 PRO questionnaires to
  retain for stage 2
• Analysis
• Prevalence of each possible symptom in the Top
  Three Most Noticed Symptoms in the last week
  (as asked by the Symptom Representation
  Questionnaire).
• Summary statistics and histograms describing the
  frequency distributions of items and domain
  scores for each symptom at baseline
• Changes from baseline in item and domain scores
• Data
• 31 patients who completed QOL questionnaires at
  baseline (pre-C1) AND (pre-C2 (AND/OR) pre-C3)
  by Jul 09

24
Results Top 10 Symptoms of the Three Most
Noticed Symptoms in the last week at baseline
Rank Symptom No. who nominated this symptom in her Top 3 (n31)
1 Fatigue 17
2 Pain - general 11
3 Abdominal bloating 10
4 Sleep disturbance 9
5 Nausea and vomiting 8
6 Appetite 7
7 Shortness of breath 6
8 Bowel disturbances (including constipation) 6
9 Pain - abdominal 5
10 Urinary problems 3
25
Coverage of Top 10 symptoms by candidate
questionnaires
Symptom SRQ Pt DATA FACT-O FOSI QLQ-C30 QLQ-OV28
Fatigue 2 1 1 1 3 -
Pain - general 1 1 1 1 2 -
Abdominal bloating 1 1 1 1 - 1
Sleep disturbance 1 1 1 - 1 -
Nausea and vomiting 2 2 2 2 2 -
Appetite 1 2 1 - 1 1
Shortness of breath 1 1 - - 1 -
Bowel disturbances 1 3 1 - 2 3
Pain - abdominal - 1 1 1 - 1
Urinary problems 1 1 - - - 1
items covering Top 10 12 15 9 6 12 9
Total items in qaire 24 18 39 8 30 28
50 83 23 75 40 32
26
Detailed example of overlap Pain - general
Qaire Item stem Response scale Scoring
SRQ Pain 0 did not have the symptom 10 as bad as I can imagine Single item
Pt-DATA Pain (all or anywhere) 0 no trouble at all 10 worst I can imagine Single item
FACT-O I have pain 0 Not at all 1 A little bit 2 Somewhat 3 Quite a bit 4 Very much One of 7 items in the Physical Wellbeing scale of the FACT One of 8 items in the FOSI
QLQ-C30 Have you had pain? 1 Not at all 2 A little 3 Quite a bit 4 Very much These two items form the pain scale of the QLQ-C30
QLQ-C30 Did pain interfere with your daily activities? 1 Not at all 2 A little 3 Quite a bit 4 Very much These two items form the pain scale of the QLQ-C30
27
Results at baseline (pre-cycle 1)

• For each of the top 10 symptoms
• we compared distributions and summary statistics
  for similar items
• No major ceiling or floor effects
• Similar distributions for similar items
• Nothing to choose between questionnaires

28
Change at Cycles 2 and 3

• For each of the top 10 symptoms
• we compared distributions and summary statistics
  of change scores on similar items
• Mean change 0 with large SD reflecting
  improvements in some women and deteriorations
  in others
• Comparable results across qaires
• Nothing to choose between questionnaires

29
Decisions

• Retain modified Pt DATA Form Ovarian to measure
  key symptoms
• Enhance coverage of the Top 10
• Allow measurement of both current status and
  change
• Modifications by developer and OSBS investigators
  ? OSBS Recent Status Form (after each cycle)
  OSBS Change Form (after every 2nd cycle)
• Develop a separate Side Effects Form for net
  clinical benefit
• Retain FACT-O (including FOSI) to measure QOL
• FACT-O
• Has fewer items 39 (incl. 8 for FOSI) vs. 58 in
  the QLQ-C30/Ov28
• Provides summary scores overall QOL, Trial
  Outcome Index, FOSI
• Overall QOL based on all items
• QLQ-C30
• 22 sub scales
• Duplication of single-item symptoms with Pt DATA
  Form
• Global QOL based on 2 items

30
Prognostic Modelvariables

• No. of lines of therapy
• Performance status
• Volume of disease
• Sites of disease
• CA125 velocity
• LDH Hb Albumin Platelets
• Inflammatory markers
• Grade histological subtype

31
Platinum Resistant Ovarian CancerPFS
32
Platinum Resistant Ovarian CancerOS
33
Platinum Resistant Ovarian CancerHypothetical
Risk GroupsPFS
34
Platinum Resistant Ovarian CancerHypothetical
Risk GroupsOS
35
Discussion Points

• Comments and questions of study and design-
  relatively fluid at present
• Comments on circulated CRFs
• Which groups will join stage 2 study and when ?
  Feasibility Time Frame Trials
• Translations
• Funding arrangements in different groups
• ECRFs and scanning of HRQOL forms