CONHIV14 - HIV-related Infections, Co-infections and Cancers, etc. – Part three - page 3

P
REVALENCE
AND
P
REDICTORS OF MALIGNANCIES
IN
A
P
OLYCENTRIC
C
OHORT OF
HIV
PATIENTS
FROM
I
TALY
E. Mazzotta
1
, M. Tontodonati
1
, C. Gabrielli
2
, S. Mazzocato
3
, M. Mazzetti
4
, K. Falasca
5
, G. Cenderello
6
,
J. Vecchiet
5
, F. Barchiesi
3
, D. Francisci
2
, G. Parruti
1
1
Infectious Diseases Unit, Pescara General Hospital, Pescara, Italy
2
Infectious Diseases Clinic, Università degli Studi di Perugia, Perugia, Italy
3
Infectious Diseases Clinic, Department of Biomedical Sciences, Università Politecnica della Marche, Ancona, Italy
4
Infectious Diseases Unit, AOU Careggi, Firenze, Italy
5
Infectious Diseases Clinic, Università “G. D’Annunzio”, Chieti, Italy
6
Infectious Diseases Unit, Galliera Hospital, Genova, Italy
;
I
NTRODUCTION
HIV infected patients have an higher risk of developing cancer than the general population. Kaposi's sarcoma (KS), non-Hodgkin's lymphoma (NHL),
primary CNS lymphoma (PCL) and invasive cervical cancers are considered AIDS defining. An increased incidence in recent years, however, has been
reported also for other malignancies after the introduction of cART.
M
ETHODS
We performed a retrospective multicentric evaluation of all HIV infected patients with both AIDS
and non-AIDS defining neoplasms at six Infectious Disease Units spread throughout Italy since
1991 through 2013. We also evaluated prevalence of tumors. Cases were compared with equal
number of controls without neoplasia followed at the same Institutions, matched for length of
HIV infection. All statistical analyses were performed using Stata 10.1 package.
R
ESULTS
Since 1991, 339 consecutive cases of malignancy were collected
from the six convening Centers, including approximately an equal
proportion of AIDS (51.6%) and non-AIDS defining tumors. Mean
prevalence of tumors among Centers was 8.3% (r. 6.1% - 9.6%).
Mean age at tumor diagnosis was significantly lower than in
controls (42.6±11.0 vs 46.8±10.6 years, respectively, p<0.0001).
As to risk factors for HIV infection, approximately ¼ (26.1%) of
patients were drug abusers, in equal proportion as in controls. A
remarkable higher proportion of cancer patients had CD4 T-cell
counts <200 c/mmc at time of diagnosis (45.2% vs 13.3%,
p<0.0001). Seventy percent of tumors occurred in males; 52.8%
of tumor patients were diagnosed with AIDS before and 19.0% at
the time of tumor diagnosis. Ninety (28.1%) tumor patients were
dead at the time of data collection, a much higher proportion than
among cases (12.9%, p<0.0001). Deaths among non-AIDS
(20.8%) and AIDS defining tumor patients (35.0%) were
significantly different (p=0.005). Predictors of AIDS defining
tumors at the time of data collection were: male sex (57.9% vs
40.6%, p=0.004), cd4 Tcell counts <200 c/mmc (63.6% vs 44.1%,
40.6%, p=0.004), cd4 Tcell counts <200 c/mmc (63.6% vs 44.1%,
p<0.0001), whereas being cART treated at the time of tumor
diagnosis was protective (38.0% vs 68.0%, p<0.0001). In the final
multivariate model of logistic regression, male sex (OR=2.0,
p=0.03) and not being cART treated (OR=2.5, p=0.001) held as
independent predictors.
C
ONCLUSIONS
Our retrospection revealed a considerably high proportion of non
AIDS defining tumors, apparently at rise in recent years. We
registered high prevalence of tumors in each Center. Being
untreated was associated with AIDS defining tumors: once more
prevention of late presentation appeared the way to avoid worse
prognosis in this setting.
R
EFERENCES
1. Centers for Disease Control and Prevention. 1993 revised classification system for HIV infection and expanded surveillance
i
l
i
i
l
i i
i
i
i
ill
case definition for AIDS among adoles-
i i i
l
cents and adults. MMWR , 1992
l
2.Engels EA, Pfeiffer RM, Goedert JJ, et al. Trends in cancer risk among people with AIDS in the United States 1980
l
i
l
i
i
l
i
i
i
-2002. AIDS. 2006;20(12):1645-54
3.Silverberg MJ, Chao C, Leyden WA, et al. HIV infection and the risk of cancers with and without a known infectious cause. A
il
l
i
i
i
i
i
i
i
IDS. 2009;23(17):2337-45.
Tab 1. Main features of Cases and Controls
Tab 2. Predictors of AIDS definig : univariate and multivariate analyses
Cases
(339)
Controls
(260)
p value
Sex, male, %
70.4
70.4
0.9
Mean age, y (SD)
42.6 (11.0)
46.8 (10.6)
<0.001
Drug addiction, %
24.5
28.8
0.2
AIDS, %
71.8
65.1
0.1
Smoke, %
59.3
54.5
0.3
Death, %
28.1
12.9
<0.001
Fig 1. Kind of tumors
AIDS defining tumor
Predictors
Univariate analyses
p
Logistic regression (obs=295)
OR [95% C.I.]
p
Sex, male vs female (%)
57.9% vs 40.6%,
40.6%,
p=0.004
2.0 [1.1
2.0 [1.1
-
-
3.5]
3.5]
0.03
Age at tumor diagnosis, mean (y)
42.5 vs 42.6
0.9
0.8
LyCD4 < 200 cell/mmc (%)
63.6% vs 44.1%,
p<0.0001
0.2
cART naïve (%)
68.0% vs 38.0%,
p<0.0001
2.5 [1.5—4.5]
0.001
HCV coinfection (%)
43.2 vs 54.9
0.07
0.5
48.4%
51.6%
1,2 4,5,6,7,8,9,10,11,12
Powered by FlippingBook