Malignant Outcomes in the CPCRA

1
Malignant Outcomes in the CPCRA

• An Historical Overview
• And
• A Look to the Future

2
Opportunistic Disease Events in the Six Months
Prior to Death
PCP
849
473
MAC
Wasting Syndrome
466
Bacterial Pneumonia
452
441
CMV
Invasive Candidiasis
406
KS-Non-Visceral
314
AIDS Dementia Complex
191
154
TB
Lymphoma
125
Toxoplasmosis
125
Cryptococcosis
112
KS-Visceral
82
Cryptosporidiosis
70
Herpes Zoster
67
52
PML
Percent
N 1883 patients from cross-study database
7/95
3
Relative Risk of Selected Events by Gender(based
on 15 months mean follow-up of cohort)
Adjusted RR WomenMen
P-value

• Women were at increased 1.38 .01
• risk for bacterial pneumonia
• Women were at decreased 0.16 lt.05
• risk for Kaposis sarcoma
• Women were at decreased 0.54 lt.05
• risk for oral hairy leukoplakia
• Disease progression including 0.97 .72
• death was equivalent
• Women were at increased risk 1.33 .01
• for death

Obtained from a proportional hazards model with
stratification by center and covariates
corresponding to age, race, history of IDU, CD4
count, Karnofsky score, disease progression
history, and use of antiretroviral drugs and PCP
prophylaxis.
Melnick et al, JAMA 2721915-1921, 1994
4
Risk of Death for AIDS Events
6 Mo. Cum. Mort.
RR of Death
No. Pts.
Event

• PML 20 84.4 18.3
• Lymphoma 57 70.8 8.1
• Visceral KS 50 49.3 4.9
• ADC gt stage 2 97 67.3 4.6
• MAC 276 43.8 2.5
• Wasting 197 41.7 2.5
• PCP 378 32.7 2.4
• CMV disease 289 41.0 2.4
• Candidiasis (inv.) 234 24.0 1.7

N 3382 patients from cross-study database
(Neaton et al, Statistics in Medicine, 1994)
7/95
5
AIDS-related Malignanciesin the HAART Era

• Analysis performed to compare risks for ARM in
  pre- vs post-HAART setting
• NuCombo (4/92) randomized pts to AZT or AZT/ddI
  or AZT/ddC
• NvR (1/97) randomized pts to either NLF or RTV
  added to background NRTIs selected by the
  physician
• ARM defined as KS /- visceral involvement, NHL
  or invasive cervical cancer
• Observations censored as of the common closing
  date of trial
• NuCombo11/30/1995
• NvR 12/31/01
• Observations available on 1860 individuals with
  median 39.7 mos f/u

6
Baseline Characteristics of Patients in NuCombo
(007) and NvR (042)
NuCombo
NvR
AZTMonotherapy
AZTddI/ddC
HAART

• No. patients 369 730 761
• Female () 7.0 8.8 16.8
• Race ()Latino 9.2 9.7 10.2Black 32.8 33.2
  42.8White 55.6 55.3 40.9
• IDU 23.2 22.9 18.7
• Male same-sex 72 75.5 58.0contact ()

7
Baseline Characteristics of Patients in NuCombo
(007) and NvR (042) (cont.)
NuCombo
NvR
AZTMonotherapy
AZTddI/ddC
HAART

• No. patients 369 730 761
• Prev. OI/OM () 36.0 37.5 53.9
• Age (mean years) 37.9 37.3 39.0
• Weight (lbs.) 161.8 160.1 154.4
• CD4 (mean cells) 117.2 120.1 57.6
• Median Follow-up (months) 34.1 34.2 52.2

8
Number of Events by Treatment Group
AZT ddI/ddC
HAART
Event
N
Pct
N
Pct

• KS (without visceral involvement) 35 9.5 59 8.1 17
  2.2
• KS (with visceral involvement) 15 4.1 22 3.0 6 0.8
  
• Lymphoma 18 4.9 19 2.6 10 1.3
• Cervical Cancer 0 0.0 0 0.0 1 0.1
• Any KS or Lymphoma 60 16.3 87 11.9 31 4.1

21 NuCombo and 2 NvR subjects had two events
Chi square tests for non-homogeneity between
columns for each event were statistically
significant (plt0.002) except cervical ca
9
Rate per Hundred Person-Years of AIDS-Related
Malignancies by Gender
Men
Women
Overall

• AZT only 343 60 9.2 26 0 0 369 60 8.6
• AZTddI/ddC 666 86 6.4 64 1 0.8 730 87 5.9
• HAART 633 29 1.3 128 3 0.5 761 32 1.2
• Total 1642 175 4.2 218 4 0.5 1860 179 3.7

10
Relative Risk from Cox Proportional Hazards
Regressionin Men Adjusted for Baseline CD4 and
Stratified by Unit
HAART Versus AZT Monotherapy
HAART Versus AZT ddI/ddC
CombinedEvent Rateper 100 Person-years
TotalNumberof Events
RelativeRisk
RelativeRisk
Event
95 CI
95 CI

• Kaposis Sarcoma(non-visceral) 109 2.6 0.14 (0.08
  ,0.25) 0.11 (0.06,0.21)
• Kaposis Sarcoma(with visceralinvolvement) 43 1.
  0 0.08 (0.03,0.24) 0.06 (0.02,0.19)
• Lymphoma 46 1.1 0.26 (0.11,0.62) 0.14 (0.06,0.32)
• Any KS or 175 4.2 0.16 (0.10,0.26) 0.11 (0.07,0.18
  ) Lymphoma

11
Event Rates (95 Confidence Intervals) for Men by
Cohort per Hundred Person-Years Follow-up
NuCombo
NvR
AZTMonotherapy
AZTddI/ddC
HAART

• Kaposis Sarcoma 5.3 (3.7, 7.4) 4.3 (2.0,
  5.7) 0.7 (0.4, 1.1)(non-visceral)
• Kaposis Sarcoma 2.2 (1.2, 3.6) 1.6 (1.0,
  2.4) 0.3 (9.1, 0.7)(visceral involvement)
• Lymphoma 2.6 (2.0, 4.1) 1.4 (0.8, 2.2) 0.4 (0.2,
  0.8)
• Any Kaposis Sarcoma 9.2 (7.1, 12.) 6.4 (5.1,
  8.0) 1.3 (0.9, 1.9)or Lymphoma

12
Time To AIDS Related Malignancy in Men
13
Relative Risks (95 Confidence Intervals) for
Important Predictors of AIDS-Related Malignancies
in Men by Cohort Adjusted for Baseline CD4 and
Stratified by Unit
NuCombo
NvR
AZT Monotherapy
AZTddI/ddC
HAART
Model/Event

• Multivariate ModelAge 0.99 (0.95,
  1.02) 0.98 (0.95, 1.01) 1.00 (0.95, 1.05)Race
  (White vs. 1.27 (0.69, 2.36) 1.20 (0.71,
  2.03) 2.55 (0.95, 6.85) Non-White)MSM 1.11 (0
  .51, 2.39) 1.87 (0.85, 4.14) 2.25 (0.80, 6.34)
• Univariate ModelAge 0.99 (0.95,
  1.02) 0.98 (0.95, 1.01) 1.01 (0.97, 1.06)Race
  (White vs. 1.26 (0.69, 2.35) 1.36 (0.82,
  2.25) 2.85 (1.14, 7.12) Non-White)MSM 1.18 (0
  .51, 2.39) 2.13 (1.00, 2.13) 2.89 (1.08, 7.75)

14
Aids-Related Malignanciesin the HAART Era

• Despite a significantly lower CD4 cell count in
  the NvR cohort, the rate of development of ARM is
  markedly decreased in the HAART era
• Women across both time periods are at decreased
  risk compared to men to develop ARM
• ARM are reduced in the HAART cohort in all
  malignancies analyzed compared to AZT monotherapy
  or to two NRTIs
• There is no evidence that ARM are related to age
  race and MSM exposure trend towards being
  important predictors

15
Non-AIDS-defining Malignanciesin the CPCRA

• CPCRA Form 885 launched March 2000 to collect
  DIAGNOSIS OF CANCER (excluding lymphoma, Kaposis
  sarcoma and invasive cervical cancer) in FIRST,
  MDR and LTM Naïve Cohort
• Source documentation autopsy, death
  certificate or note, hospital discharge
  summary, pathology report, report of results of
  procedure giving rise to diagnosis in medical
  record
• Specify the site and type of cancer
• Resulting diagnosis confirmed or ruled out by
  SDMC

16
Baseline Characteristics of Patients inFIRST,
MDR, and the LTM Naïve Cohort
LTMNaïve Cohort
MDR HIV
FIRST

• No. patients 274 1373 786
• Female () 9.1 20.7 23.0
• Race ()Latino 15.3 16.8 14.4Black 38.0 53.
  9 47.7White 44.9 26.6 35.1
• IDU 10.7 14.6 16.8
• Male same-sex 73.5 60.6 71.4contact ( of
  males)

17
Baseline Characteristics of Patients inFIRST,
MDR, and the LTM Naïve Cohort
LTMNaïve Cohort
MDR HIV
FIRST

• No. patients 274 1373 786
• Prev. OI/OM () 56.6 29.6 10.3
• Age (mean years) 44.3 38.3 38.4
• Body Mass Index (kg/m2) 24.8 24.5 26.3
• CD4 (mean cells) 182.4 210.4 493.8
• Hepatitis B () 1.5 3.6 5.6
• Hepatitis C () 7.32 17.9 17.0
• Malignancy () 4.0 1.0 0.6
  (non-HIV-defining)

18
Rates per Hundred Years of Follow-upof Selected
Events in MDR, FIRST, and LTM Naïve Cohort by
Gender
Male
Female
Overall

• Diabetes requiring .52 (.31, .80) .31 (1.06,
  .91) .48 (.30, .72)drug treatment
• Lactic acidosis .36 (.20, .53) .51 (.17,
  1.18) .39 (.23, .61)
• Malignancy .64 (.40, .97) .79 (.32,
  1.6) .67 (.45, .96)(non-AIDS defining)

19
50 Non-AIDS DefiningMalignancies Reported

• Basal cell carcinoma 12
• Lung Cancer 9
• Anal squamous cell ca 6
• Squamous cell skin 3
• SCC unknown primary 3
• Head and Neck 3
• Breast 2
• Prostate 2
• Renal cell carcinoma 2

• One of Each of the Following
• Adenocarcinoma unknown 1º
• Bladder
• Colon
• Esophagus
• Hairy cell leukemia
• Hodgkins disease
• Leiomyosarcoma
• Vulvar

20
Non-AIDS defining Malignancies

• The rate of non-AIDS defining malignancies seems
  comparable to the development of some metabolic
  complications
• Skin malignancies are common
• In contrast to ARM, women are equally impacted
• Response to conventional Rx and survival data are
  currently unavailable

21
Future ResearchPossibilities in Malignancies

• Hepatocellular carcinoma in co-infected patients
• Hepatocellular and prostate cancer in
  androgen/anabolic users
• Effect of treatment interruptions on malignant
  outcomes
• Impact of IL-2 on development of malignancies
• Impact of CD4 rise and PI vs non-PI regimen in
  KS Rx
• Response to therapy of non-ARM in HIV patients
• Impact of chemo/XRT regimens on CD4, HIV RNA
• Timing of HAART in newly diagnosed HIV with NHL
• High dose methotrexate in primary CNS lymphoma
• Anal squamous cell carcinoma treatment and
  prophylaxis

22
HELP WANTED

• San Francisco General Hospital
• University of California San Francisco
• SEEKS TWO FACULTY ONCOLOGISTS ASAP!!!