Assessing Response and Progression in Ovarian Cancer Clinical trials RECIST 1.1 and CA125 Criteria

1
Assessing Response and Progression in Ovarian
Cancer Clinical trialsRECIST 1.1 and CA125
Criteria

• E.A. Eisenhauer
• GCIG Ovarian cancer clinical trials planning
  meeting
• May 29, 2009

2
Outline

• Response /progression Criteria
• RECIST 1.0/1.1
• CA125 criteria
• Major uses in ovarian cancer trials
• Issues / questions

3
RECIST

• Response evaluation criteria in solid tumours
• Anatomically based
• Major uses in solid tumour trials
• Phase II objective response, stable disease
  rates
• Phase III PFS
• Several issues with initial RECIST, for e.g.
• Assessment of PD in non-measurable disease
• Lymph node assessment
• Number of lesions
• RECIST 1.1 (EJC, Jan 2009) addresses all and
  more. Changes based on evidence from trial data
  and simulation work.

4
What HAS changed in RECIST 1.1
RECIST 1.0 RECIST 1.1
Measuring tumor burden 10 targets 5 per organ For response 5 targets (2 per organ)
Lymph node Measure long axis as for other lesions. Silent on normal size Measure short axis. Define normal size.
Progression definition 20 increase in sum 20 increase and at least 5 mm absolute increase
Non-measurable disease PD must be unequivocal Expanded definition to convey impact on overall burden of disease. Examples.
Confirmation of objective response required Required when response primary endpointbut not otherwise
New lesions -- New section which includes comment on FDG PET interpretation
5
GCIG CA125 criteria

• Developed by GCIG to complement RECIST criteria
• Response criteria1
• For use in phase II studies in relapsed disease
• Evaluable pts must have baseline CA125 gt 2 x ULN
• CA125 response 50 decrease confirmed gt 28 days
• Date response date of first 50 fall
• Progression Criteria2
• For use in front-line setting to complement
  objective PD
• CA125 PD Double of UNL (or nadir if gt UNL)
  confirmed gt 7 days
• Date CA125 PD date of first doubling
• Date overall PD earliest date of CA125 or
  objective PD
• Exception recent surgery or intraperitoneal
  procedure

1. Rustin GJ, et al J Natl Cancer Inst. 2004
   96487-8
2. Vergote I, et al. J Natl Cancer Inst. 2000,
   921534-5

6
Primary Endpoints Front-line studies

• Baden Baden meeting 2004
• Advanced OVCA first-line Both PFS and OS are
  important endpoints to understand the full impact
  of any new treatment. Thus either may be
  designated as the primary endpoint. Regardless of
  which is selected, the study should be powered so
  both PFS and OS can be appropriately evaluated
• Good evidence that PFS is surrogate for OS, and
  acceptable endpoint for front-line studies

7
HR of PFS vs OS Front-Line Trials
CAP vs CA (GOG 47)
T vs P (GOG132b)
8
Issues with combined use of CA125 and Objective
PD in front-line studies

• Data collection more complex
• Timing of CA125 measurement very important to
  avoid bias
• If PD by one method, do we still need to collect
  PD by other?
• How many PD events are found solely by CA125?
• Do we count CA125 PD if patients still on
  front-line therapy? (some Groups do, some do not)

9
OV.16 Progression Events
Arm 1 N 409 Arm 2 N 410 Total
No. pts with PD 329 321 650
Objective PD only 117 112 229
CA125 PD only 24 17 41
Both Objective and CA125 PD 178 181 359
Death without PD 10 11 21
10
OV.16 PFS (GCIG definition Objective and CA125
-- whichever is first)
Arm 1 median 14.6 mo Arm 2 median 16.2 mo Hazard
ratio 1.10 p 0.25 Adjusted p 0.23
11
OV.16 PFS Objective PD only
Arm 1 median 17.5 mo Arm 2 median 17.9 mo Hazard
ratio 1.03 p 0.69
12
CA125 PD definition

• Does including CA125 as part of PD definition add
  value or just complexity?
• Majority of pts with elevated CA125 have imaging
  and are found to have objective PD
• i.e. would it be enough to measure CA125
  routinely BUT to consider PD based on objective
  findings only? Main advantage trial conduct
  simplified

Need data from other front-line trials which used
GCIG definition PD to determine if value added by
this composite endpoint
13
Primary Endpoints Recurrent Disease Studies

• Baden Baden meeting 2004
• Phase II screening trials
• Response using RECIST criteria or GCIG CA125
  response definition -- to be specified in
  protocol
• Phase III second-line
• Symptom control/Quality of life (for early
  relapse) and overall survival (for late relapse)
  are preferred primary endpoints, although PFS
  should still be used in assessing new treatments.

14
Issues recurrent disease studies (1)

• Phase II screening trials
• CA125 RR is usually higher than objective RR
  does this mean anything?
• Are drugs with CA125 responses but no objective
  responses active? Have any been identified and
  tested in phase III?
• Is CA125 RR value added in drug development?
  (there is no doubt CA125 is useful in clinical
  management but that is another question)
• Are other functional/molecular imaging endpoints
  of value?
• Is non-progression rate (CR PR SD) more
  meaningful than response rate?
• Need DATA to answer all these questions!

15
Issues recurrent disease studies (2)

• Phase III trials
• Although Baden Baden endpoint recommendations are
  symptom benefit or OS, PFS is often used in these
  trials as primary endpoint.
• Does PFS prolongation of 1-2 mo have any
  meaning for patients in situation of recurrent
  disease if NOT accompanied by either OS or
  symptom improvement?
• Is PFS a surrogate for OS in recurrent disease?

16
HR of PFS vs OS Second line combination trials
17
Issues recurrent disease studies (2)

• Phase III trials
• Although Baden Baden endpoint recommendations are
  symptom benefit or OS, PFS is often used in these
  trials as primary endpoint.
• Does PFS prolongation of 1-2 mo have any
  meaning for patients in situation of recurrent
  disease if NOT accompanied by either OS or
  symptom improvement?
• Is PFS a surrogate for OS in recurrent disease?
• Is PFS a surrogate for symptom improvement?
• If so isnt it easier and more meaningful to
  measure symptom benefit?
• If so, shouldnt trials with PFS improvement show
  symptom benefit? Do they??

Need DATA to answer all these questions!
18
Extra slides
19
Relationship between PFS and OSRecent
front-line RCTs in OVCA
Trial (experimental vs standard) HR PFS HR OS
GOG 47 CAP vs CA 0.715 0.936
GOG 158 TC vs TP 0.88 0.84
GOG 132 TP vs P 1.06 0.99
GOG 132 T vs P 1.41 1.15
OV10 TP vs CP 0.74 0.73
GOG 111 TP vs CP 0.73 0.75
ICON3 TC vs C/CAP 0.93 0.98
AGO TEC vs TC 0.95 0.93
GOG 172 IP TP vs IV TP 0.77 0.73
AGO TC vs TP 1.05 1.045
AGO TC topo vs TC 0.97 1.01
20
HR of PFS vs OS in second line trials
PFS OS
ICON4 0.76 0.82
OVAR 2.5 0.72 0.96
Gonzales-Martin (GEICO) 0.54 0.31
Cantu 0.60 0.58
Bolis 0.60 0.80