A NotSoSilent Killer

1
A Not-So-Silent Killer

• Symptoms of Ovarian Cancer

Barbara A Goff, MDDepartment of Obstetrics and
GynecologyDirector, Division of Gynecologic
OncologyUniversity of Washington School of
Medicine
2
Survival Epithelial Ovarian Cancer

• Disease Stage Presentation Survival
• Early 2030 7090Advanced 7080 2030

3
Screening Challenges Ovarian Cancer

• No precursor or in situ lesion
• Annual incidence 40/100,000 women
• Major surgical procedure required for diagnosis
• Therefore, morbidity and mortality of false
  positive screening must not outweigh benefits of
  early detection

4
CA125 Tumor Marker Ovarian Cancer Screening

• Elevated in 80 of epithelial cancers and 1 of
  normal controls
• Studies from Europe have shown CA125 can detect
  ovarian cancer in asympto-matic women
• No impact on overall mortality
• Only 50 of women with Stage I disease have
  elevations in CA125

5
CA125 Tumor Marker Ovarian Cancer Screening

• 22,000 post-menopausal women
• CA125 and, if elevated, TVS
• 41 women underwent laparotomy
• 11 ovarian cancers detected (4 early and 7 late)
• 8 women with negative screens developed ovarian
  cancer

Jacobs et al. BMJ 1993
6
CA125 Tumor Marker Ovarian Cancer Screening

• 200,000 being enrolled in screening trial in UK
  to be completed in 2011
• Serial CA125 with TVS if abnormal vs controls
• Report of first 13,582 patients (6,682
  randomized to screening) from 1995-2000
• Only 3 ovarian cancers detected

Menon et al. JCO 2005
7
Screening with TVS Ovarian Cancer

• 14,469 asymptomatic women
• 57,214 scans over 10 years
• 180 women underwent surgery
• 17 cancers detected, but only 11 invasive
  epithelial cancers Stage I 5 Stage
  II 3 Stage III 3
• 5,200 TVS and 16 surgeries/case of invasive Ca

Van Nagell et al. Gyn Onc 2000
8
Multimodality ScreeningPLCO Screening Trial

• Enrolled 78,232 women from 19932001
• Age 5574
• Randomized to annual TVS/CA125 vs routine care
• Data from baseline analysis of 28,816 women
  randomized to screening CA125 ? in 1.5 (PPV
  for cancer 3.7) TVS abnormal in 4.7 (PPV 1.0)

9
Multimodality ScreeningBaseline Data PLCO
Screening Trial

• 1,740/28,816 had either abnormal CA125 or TVS
  34/28,816 had both abnormal tests
• 570 women underwent surgery
• 20 invasive cancers identified
• PPV of both tests abnormal 23.5, but 12/20
  cancers would have been missed if both tests
  required to be abnormal

10
PLCO TrialResults From 4 Yrs Annual Screening

• Results from screening arm only 98 cancers
  detected 62 (63) initial scrn 22 (23)
  interval scrn 14 (14) not P/U 77 detected
  by TVS were Stage I/II 90 detected by CA125
  were Stage III/IV (impact on mortality awaiting
  completion of trial and comparison to control
  group)
• PPV over 4 years CA125 2.03.7 TVS 0.50.99

11
High-Risk Populations Ovarian Cancer Screening

• 4,526 women with family or personal history of
  cancer and normal initial TVS
• TVS every 6 months
• 10 ovarian, fallopian tube and primary peritoneal
  cancer diagnoses in advanced stages after normal
  TVS
• Authors concluded that TVS has limited value as
  independent screening modality

Fishman et al. Am J Ob/Gyn 2005.
12
Screening High-Risk Populations PLCO Trial
Baseline 3 Years

• Patients stratified by risk
• Average No family history Moderate One 1
  relative with breast Ca High Ovarian
  cancer 2 relatives with breast Ca
  Personal hx breast Ca
• PPV of abnormal screening test not signifi-cantly
  different for various risk groups

13
Results Screening Exam/Risk Grp PLCO Trial
Baseline 3 Years

• Average
  Moderate High
  (n22,687) (n2,572)
  (n2,163)
• Total screen exams 82,626 9,417 7,903
• Normal 78,531 8,953 7,520
• Pos elevated or abnl 4,095 (5.0) 464 (4.9) 383 (
  4.8) Pos ? biopsy 21 26 27 Invasive Ov
  Ca 30 6 5 Invas tub/peritoneal 8 1 0
  LMP tumors 9 1 3
• PPV of screening test Invasive Ov
  Ca 0.7 1.3 1.6 Plus tub/peritoneal 0.9 1.5
  1.6 Plus LMP tumors 1.1 1.7 2.3

14
Serum Protein Markers Early Detection of Ovarian
Cancer

• Microarray analysis used to identify significant
  differences between cases and controls in 6
  markers Leptin Osteopotin --CA125
  ILGF-II Prolactin --MIF
• Training set of 181 controls and 113 OC patients
• In a blind evaluation of 181 controls and 43
  ovarian cancer patients, a combination of 4
  proteins yielded a Sensitivity of 95.3
  Specificity of 99.4

Visintin, Cancer Res 2008
15
RecommendationsCurrent Ovarian Cancer Screening

• 1994 NIH Consensus PanelObtain family history
  Offer screening to those with 2 or more affected
  family members
• ACOGShould not offer screening to general
  population
• US Preventive Services Task ForceOvarian cancer
  screeningGrade D Fair evidence to recommend
  exclusion from periodic health exam

16
High-Risk Women Ovarian Cancer Recommendations

• BSO when childbearing complete
• Risk of primary peritoneal cancer remains
• For those not ready or unwilling to undergo BSO,
  enrollment into clinical trials for screening
  should be encouraged

17
Question Ovarian Cancer

• Currently, screening not recommended for general
  population
• 90 of ovarian cancers occur in women without a
  family history
• Is there a way to make an earlier diagnosis in
  these women?
• Historically, ovarian cancer called the silent
  killer

18
Previous Studies Ovarian Cancer Diagnosis

• Results of a national survey
• To conduct large national survey of women with
  ovarian cancer
• To evaluate preoperative symptoms which patients
  experience
• To evaluate potential causes in delayed diagnosis

Goff, Muntz, Mandel, Melancon. Cancer 2000.
19
Previous Studies Symptoms of Ovarian Cancer

• 1,725 women with ovarian cancer surveyed
• 95 had symptoms Abdominal/GI the most
  common Pelvic symptoms least common
• 89 of women with Stage I/II disease had symptoms

Goff et al. Cancer 2000.
20
Duration of Symptoms Ovarian Cancer Diagnosis

• Months
• 02 30 36 35 712 20 gt12 15

21
Results Ovarian Cancer Diagnosis

• Initial diagnosis Irritable
  bowel 15 Nothing 13 Stress 12 Gastr
  itis 9 Constipation 6 Depression 6
  Other 47
• Treated for another condition 30

22
Conclusions Ovarian Cancer Diagnosis

• Majority of women with ovarian cancer do have
  symptoms
• Majority of women with early stage disease do
  have symptoms
• Women often unaware of what consti-tutes normal
  physiologic changes with aging as opposed to
  pathologic changes

23
Conclusions Ovarian Cancer Diagnosis

• Delay in diagnosis is common
• Patient-related factors
• Physician-related factors

24
Methods Symptoms of Ovarian Cancer Study

• 168 cases ovarian cancer
• 251 controls
• Interviewed after diagnosis (mean time4.7
  months) and asked about Symptoms Medications
  Lactose intolerance

Olson et al. Obstet Gynecol 200198(2)212217.
25
Ovarian Cancer Symptoms Symptoms of Ovarian
Cancer Study

• Symptom Cases Controls OR (95
  CI)
• Any 156 (93) 106 (42) Bloating 119 (71) 22 (9
  ) 25.3 (15.6-40.9)Abdom. pain 87 (52) 37 (15)
  6.2 (4.0-9.6)Fatigue 72 (43) 40 (16) 3.9 (2.5-6
  .1)Urin. Freq. 55 (33) 31 (12) 3.5 (2.2-5.7)La
  ck appetite 34 (20) 7 (3) 8.8 (4.3-18.2)Constip
  ation 36 (21) 18 (7) 3.5 (2.0-6.3)

Olson et al. Obstet Gynecol 200198(2)212217.
26
Symptoms Symptoms of Ovarian Cancer Study

• Concluded that women with ovarian cancer do have
  symptoms Present in 89 of women with early
  stage disease
• Study weaknesses Small numbers Recall
  bias Conducted only in New York city

Olson et al. Obstet Gynecol 200198(2)212217.
27
Frequency of Ovarian
Cancer Symptoms in Women Presenting to Primary
Care Clinics
Barbara Goff, MDLynn Mandel, PhDCindy Melancon,
RNHoward Muntz, MD
28
September 6, 2004


29
Study Objectives Symptoms of Ovarian Cancer

• To identify frequency, severity and duration of
  symptoms typically associated with ovarian cancer
  in a population of women presenting to primary
  care clinics
• Comparison was made to 128 women with masses who
  were surveyed about symptoms prior to surgery

Goff et al. JAMA 2004.
30
ResultsPrimary Care Clinics Symptoms of Ovarian
Cancer

• 1,709 women in primary care clinic completed
  survey General checkup 25 Mammogram 13 Prob
  lem visits 62
• 95 reported at least 1 symptom in past yr Back
  pain 60 Fatigue 52 Indigestion 37 Urinary
  symptoms 35 Constipation 33
• Median number reported symptoms was 4
• Median severity 2-3

31
ResultsPrimary Care Clinics Symptoms of Ovarian
Cancer

• Patients presenting for general checkup had fewer
  symptoms (p0.001)
• As age ?, all symptoms Less common, except
  urinary symptoms Less severe, except urinary
  symptoms

32
OR of Ovarian Ca Symptoms Women With and Without
Cancer

• Ca vs Benign Cancer vsSymptoms Ovar
  Tumor Clinic Patients
• Pelvic pain 1.8 (0.8-4.0) 2.2 (1.2-3.9)
• Abdominal pain 1.8 (0.8-4.0) 2.3 (1.2-4.4)
• Difficulty eating 2.5 (0.9-6.8) 2.5 (1.3-5.0)
• Bloating 3.5 (1.5-8.2) 3.6 (1.8-7.0)
• Abdominal size 3.0 (1.3-6.9) 7.4 (3.8-14.2)
• Urinary urgency 3.5 (1.6-8.2) 2.5 (1.3-4.8)
• Constipation 3.5 (1.5-8.1) 1.6 (0.9-3.0)
• Fatigue 1.1 (0.5-2.6) 1.4 (0.7-2.7)

Excluding patients presenting for routine
checkup or mammogram only.
33
Median Episodes Ea Sx/Mon Symptoms of Ovarian
Cancer

• Ovar Masses Primary Care
  Clin Malig Benign w/IBS w/o IBSSymptom (n44) (n
  84) (n109) (n1600) p
• Pelvic pain 24 4 2 2 .001
• Abdom pain 23 7 3 2 .017
• Bloating 30 20 4 2 .004
• Fatigue 30 28 25 8 .001
• Urinary sx 30 30 25 12 .02
• Constipation 12 30 2 2 .001
• Diarrhea 6 4 9 2 .06

34
Median Duration Ea Sx in Mos Symptoms of Ovarian
Cancer

• Ovar Masses Primary Care
  Clin Malig Benign w/IBS w/o IBSSymptom (n44) (n
  84) (n109) (n1600) p
• Pelvic pain 3 2 18 11 .06
• Abdom pain 5 3.5 12 11 .05
• Bloating 3 2 18 12 .04
• Urinary sx 3 12 12 13 .13
• Constipation 3.5 11 24 12 .001
• Diarrhea 5 3 21 12 .001
• Fatigue 3 4 21 12 .08

35
Possibilities for Earlier Detection Development
of Ovarian Cancer Symptom Index

• Prospective case-control study 149 with
  ovarian cancer 255 in OCED screening
  program 233 referred for pelvic U/S
• Evaluated symptoms Type Frequency
  Severity Duration

Goff et al. Cancer 2007.
36
Possibilities for Earlier Detection Development
of Ovarian Cancer Symptom Index

• Index considered positive Abdominal/pelvic
  pain ? Abdominal size/bloating Difficulty
  eating/feeling full
• If present lt1 year and occurred gt12 days/month

Goff et al. Cancer 2007.
37
Possibilities for Early Detection Performance of
Symptom Index

• Age 50 66.7 90.0 lt50 86.7 86.7
• Disease stage Early 56.7 Late 80.0
• Percent testing positive General population
  45/1,709 (2.6) Age 50
  8/560 (1.4) Age lt50
  33/1,102 (3.3)

38
Wednesday, June 13, 2007
39
Ovarian Cancer Can We Make the Clinical
Diagnosis Earlier?

• Evaluation of Medicare claims linked with
  SEER database Ovarian cancer patients 1,985
  Breast cancer patients 6,024 Non-cancer
  patients 10,941
• Prevalence of ICD-9 diagnosis and procedure
  codes compared among groups

Smith et al. Cancer 2005.
40
Ovarian Cancer Can We Make the Clinical
Diagnosis Earlier?

• Ovarian cancer patients more likely than
  controls to have visits for target symptoms 16
  months prior to diagnosis
• Symptoms OR (95 CI) Abdominal
  pain 6.0 (5.16.9) Abdominal swelling 30.9 (21.4
  44.8) GI complaints 2.3 (1.83.1) Pelvic
  pain 4.3 (2.86.7)

Smith et al. Cancer 2005.
41
Conclusions Symptoms of Ovarian Cancer

• Ovarian cancer causes symptoms
• Identification of symptoms by women and
  practitioners may allow for earlier diagnosis
• Until effective screening tests are developed,
  educating women and practitioners about symptoms
  is important