 |
|
|
| |
Incidence of Non-AIDS defining Malignancies in the HIV Outpatient Study (HOPS) Higher Than General Population
|
| |
| |
Incidence of 5 Non-AIDS Cancers Appear Higher in HIV-infected: CD4 nadir associated with risk
Reported by Jules Levin
Patel Pragna from the Division of HIV/AIDS Prevention, centers for Disease Control and Prevention, Atlanta, GA, reported these study results at the 11th Retrovirus Conference (Feb 8-11, 2004, San Francisco). Following is the text of her report. During the Q&A I asked her why hepatitis was not included and she said they did not find it to be a significant factor in leading to deaths; how about that !!!! I told her the CDC should be examining the rates of liver cancer and reporting this!!! So much for our CDC.
Dr Patel reported from her study findings: the incidence of five non-AIDS defining malignancies -- lung cancer, head and neck cancer, Hodgkin's disease, anorectal cancer and melanoma -- are significantly higher in HIV-infected persons than in the general population. Conversely, rates of the AIDS-defining malignancies Kaposi's sarcoma and cervical cancer have decreased, while the rate of non-Hodgkin's lymphoma has stayed the same.
Dr. Patel, based in Atlanta, and other HOPS members evaluated cancer incidence rates over an 11-year period, between 1992 and 2002.
HOPS included nine clinics in eight cities where more than 3000 HIV patients were seen annually. Also included in the analysis were close to 8000 patients seen at two Chicago HIV clinics. These data were compared with data complied for the general population and cancer registries. The numbers were adjusted for age, race, smoking, and gender.
Compared with the general population, Dr. Patel's group found that HIV-infected patients in HOPS were twice as likely to have lung cancer, 5 times more likely to have Hodgkin's disease, 10 times as likely to have anorectal cancer, and 3 times more likely to have melanoma.
For the Chicago-based patients, the risks of all five non-AIDS-defining cancers were increased in HIV patients compared with the general population. The risks of lung cancer, Hodgkin's disease, anorectal cancer, melanoma, and head and neck cancer were increased by 4-, 77-, 5-, 4-, and 10-fold, respectively.
The mean CD4 count was lower in all the cancer patients, "suggesting that immunosuppression may lead to poor tumor surveillance, allowing for the development of these cancers," Dr. Patel said. Furthermore, she said, co-infections with oncogenic viruses in the face of HIV may render patients more susceptible to developing certain cancers…for example, [human papillomavirus] is associated with anorectal cancer, and head and neck cancer, while [Epstein-Barr virus] is associated with Hodgkin's disease.
Lower nadir CD4 count was associated with risk for developing the cancers (see data below). As well, study results suggest smoking might be associated with increased risk.
HAART has improved survival of HIV-infected persons. This has led to a shift in the natural spectrum of HIV disease:
--decline in overall incidence of opportunistic infections
--unobserved medical problems: liver and kidney disease
AIDS defining cancers remain uncommon.
The objectives of this study are to describe trends in incidence of AIDS-defining malignancies:
--Kaposi's sarcoma (KS)
--Non-Hodgkin's Lymphoma (NHL)
--cervical cancer
And to describe incidence rates of non-AIDS defining malignancies in HIV-infected persons.
Studies have shown a decline in AIDS-defining malignancies since the advent of highly active antiretroviral therapy (HAART). However, the incidence of non-AIDS defining cancers among HIV-infected individuals seem to be increasing. We determined the incidence of 5 cancers among HIV Outpatient Study (HOPS) patients relative to that observed in the general population.
The HIV Outpatient Study (HOPS) is a prospective dynamic cohort of ambulatory HIV-infected patients; 9 clinics in 8 cities across the US; 3,000 persons seen annually; over 175,000 outpatient visits since 1992; data is abstracted from charts at the time of visit, then centrally compiled and analyzed.
They calculated incidence rate ratios adjusted for age, race, gender, and smoking. Age-, race-, smoking-, and gender-adjusted relative rates of 3 AIDS-defining malignancies (Kaposi's sarcoma [KS], non-Hodgkin's lymphoma [NHL], and cervical cancer) and 5 non-AIDS defining malignancies (lung, head/neck, Hodgkins disease [HD], anorectal, melanoma) and other cancers (breast, colon, prostrate) in 6,700 patients at 2 Chicago HIV clinics (Chicago HOPS patients) were compared with 20 million Cook County and 92 million Illinois cancer registry patients; and 5,400 "all other HOPS" (excluding Chicago sites) patients were compared with 334 million SEER registry patients (Surveillance, Epidemiology, and End Results Program), representative of the general population, for the years 1992 to 2002.
Smokers were defined as current or former smokers who had quit within the past 10 years or the 10 years prior to cancer diagnosis. They assigned national and local smoking rates in cancer patients to cancer registry patients and calculated adjusted incidence rate ratios using multivariate logistic regression for the 11 year period (1992-2002).
This was a retrospective chart review and abstraction of relevant information: demographic and risk data; diagnosed diseases (I'll bet HCV & HBV was not diagnosed for some patients); medication history; laboratory data. The University of Illinois at Chicago is a new site so Patel said many cancer patients are not yet enrolled in HOPS.
RESULTS
Among HOPS patients incidence rates of common cancers: breast, colon and prostrate cancers were 0.8-1.2; incidence rates were not significantly elevated in all HOPS patients.
KS and cervical cancer incidence has declined in the HAART era (1996-2002), but NHL remains the same. Incidence rates of five non-AIDS cancers (lung, head/neck, Hodgkin's, anorectal, melanoma) in HIV-infected persons are significantly higher than in the general population, despite age, race, gender, and smoking history. In the graphs Patel presented rates for cancersappeared to decline from the 1995-96 period compared to the 1997-98 period and then increased through 2001-2002, I don't know if the decline immediately after the availability of PI-HAART in 1996 is significant or not; was the immediate effect of initial HAART effective in reducing cancer rates which subsequently ijncreased as patients failed HAART or is this just an insignificant trend? Patel commented that lower mean nadir CD4 counts were seen in all cancer patients. This raises the question about when to begin therapy. I spoke with her about this. We don't have data on whether cancers may be more likely to develop if HAART is initiated below a certain CD4 level. For example, we don't know if starting HAART when CD4 counts are 350 increases risk for cancers.
CHARACTERISTICS OF HOPS PATIENTS: CANCER VS NON-CANCER
NADIR CD4 COUNT:
For HOPS patients developing these cancers these were their nadir CD4 counts:
--lung (n=32): 201
--head/neck (n=16): 103
--Hodgkin's (n=16): 181
--Anorectal (n=17): 105
--melanoma (n=11): 95
--FOR PATIENTS IN HOPS NOT DEVELOPING CANCER:
(n=5,530), CD4 nadir was 233.
SMOKERS
65% of HOPS patients not developing cancers were smokers while for patients developing cancers a higher percentage were smokers:
--lung: 100% were smokers
--head/neck: 92% were smokers
--Hodgkin's: 87%
--anorectal: 79%
--melanoma: 85%
ADJUSTED INCIDENCE RATE RATIOS of NON-AIDS CANCERS, "CHICAGO HOPS', 1992-2002
|
|
IRR*
|
95% CI
|
p-value
|
|
Lung
|
1.97
|
(1.08, 3.67)
|
0.02
|
|
Head/neck
|
7.68
|
(4.26, 13.9)
|
<0.001
|
|
Hofglins
|
27.9
|
(11.6, 87.1)
|
<0.001
|
|
Melanoma
|
6.58
|
(1.64, 26.3)
|
<0.001
|
|
Anorectal
|
102.4
|
(51.8, 200)
|
<0.001
|
*adjusted for age, race, gender, and smoking
ADJUSTED INCIDENCE RATE RATIOS (IRR) of NON-AIDS CANCERS ‘all other HOPS', 1992-2002
Among the HOPS patients, in analyses adjusted for age, race, smoking, and gender, incidence of these cancers was significantly greater than expected from SEER data:
|
|
IRR adj*
|
95% CI
|
p-value
|
|
Lung
|
2.91
|
(1.76, 4.84)
|
<0.001
|
|
Head/neck
|
0.74
|
(0.10, 5.23)
|
0.76
|
|
Hodgkin's
|
16.8
|
(8.02, 35.3)
|
<0.001
|
|
Melanoma
|
5.84
|
(2.43, 14.1)
|
<0.001
|
|
Anorectal
|
79.7
|
(35.8, 177)
|
<0.001
|
*adjusted for age, race, gender, and smoking
But she agreed this should be studied and perhaps data could be collected from HOPS to address this question but it has not been done. She recommended that HIV-infected individuals should stop smoking and should receive smoking cessation counseling and this might improve cancer rates. She also said we should increase surveillance of cancers in HIV-infected persons with risk factor data. But again, neither she nor any other CDC researcher addressed the rates or incidence of liver cancer or end srage liver disease at this conference, although numerous studies find that liver disease has become the number 1 or 2 leading cause for sickness and death in HIV.
|
|
| |
| |
|
| |
|
|
|
|
|
