HIV-related bladder cancer: A ten-year experience

Introduction

With the widespread adoption of antiretroviral therapy (ART), life expectancy among people living with HIV (PLWH) has increased markedly. ¹ Notably, the incidence of non-AIDS-defining cancers (NADCs) has risen steadily, now accounting for one quarter to one third of all deaths in this population.^2–4 Bladder cancer, a NADC, ranked 12th in global incidence among 36 cancers in 2020, with 570,000 new cases worldwide. ⁵

Data on bladder cancer in PLWH remain limited. To our knowledge, the first report originated from Uganda, where a decline in bladder cancer incidence coincided with the emergence of HIV. ⁶ Similarly, the US HIV Cancer Match Study reported a lower bladder cancer incidence in PLWH than in the general population (standardized incidence ratio [SIR] = 0.7). ⁷ However, a French series of 15 PLWH with baldder cancer suggested that HIV was associated with poorer outcomes. ⁸ A subsequent US study of 11 such patients found they presented at a younger age with only mild immunosuppression, yet their disease course resembled that of the general population. ⁹

In 2010, Chinese researchers identified two bladder cancer cases in a cohort of 3554 PLWH—the first study in China to estimate cancer incidence in this population—reporting an SIR of 4.9 and poorer outcomes. ¹⁰ Nevertheless, reports on bladder cancer in PLWH remain scarce and inconclusive regarding incidence, risk factors, pathology, and outcomes. In this preliminary study—the largest to date—we investigated the clinical characteristics, treatments, pathology, long-term survival, and HIV-related parameters in PLWH with bladder cancer.

Methods

We reviewed the medical records of patients treated at the Department of Urology, Beijing Youan Hospital, Capital Medical University, between 2013 and 2023, identifying those with both HIV and bladder cancer. Demographic, clinical, and HIV-related data were collected. For HIV-related parameters, CD4 count was measured within 2 weeks before bladder cancer diagnosis, and HIV viral load was measured within 3 months before diagnosis. Viral suppression was defined as HIV RNA <40 copies/mL.Nadir CD4 count means the lowest historical CD4 count. Duration of ART prior to cancer diagnosis was also recorded. Treatment protocols, pathology results, and long-term survival were documented.

After radical cystectomy (RC), patients received adjuvant gemcitabine-cisplatin plus tislelizumab (TGC). Treatment-related adverse events and response (objective, partial, and complete) were assessed at baseline and every 12 weeks until progression, unacceptable toxicity, or death.

Descriptive statistics characterized the cohort. Cancer-specific survival (CSS) was estimated using Kaplan–Meier analysis, and predictors were evaluated with univariate Cox proportional hazards models (SAS 9.3; significance at p < 0.05). Variables included in the univariable Cox regression analyses were selected based on clinical relevance and previously reported prognostic factors in bladder cancer. In the univariate Cox regression analyses, each variable was assessed individually without adjustment for other covariates, as the small sample size precluded multivariate modelling. The proportional hazards assumption was tested for each categorical variable by inspecting log-minus-log survival plots; all variables satisfied the assumption, as indicated by approximately parallel curves.

This study was reviewed and approved by the Ethics Committee of Beijing Youan Hospital, Capital Medical University (Approval No. [2020]035). The study was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all individual participants included in the study.

Results

Clinical characteristics

Twenty PLWH underwent treatment for bladder cancer during the study period. All were male, with a median age of 56 years (range: 28–78). Eight patients had concurrent syphilis diagnosed at the time of bladder cancer diagnosis, and three had concurrent hepatitis B virus (HBV) infection. Hypertension was the most common comorbidity (n = 3). Two patients had a >10-years smoking history; the remainder were never-smokers. Seventeen patients (85%) presented with painless gross hematuria; the remaining three were diagnosed incidentally following ultrasound. Cystoscopy was performed in all cases; tumor locations are summarized in Table 1.

OPEN IN VIEWER

Table 1.

Clinical characteristics and HIV-relevant parameters.

Case	Gender	Age	Comorbidity	Presentation	CD4 (cells/μl)	Viral load (copies/ml)	ART duration (months)	Tumor location
1	Male	51	Non Hodgkin's Lymphoma	Hematuria	527	<40	24	Single, left wall
2	Male	72	Syphilis	Hematuria	202	<40	12	Single, bladder neck
3	Male	50	None	Hematuria	459	<40	12	Multiple, Whole bladder
4	Male	59	Hepatitis B	Hematuria	227	<40	96	Multiple, trigonum
5	Male	31	None	Hematuria	759	<40	36	Single, right wall
6	Male	66	None	Hematuria	535	<40	84	Multiple, posterior, top wall
7	Male	47	Hepatitis B	Hematuria	293	<40	120	Multiple, Trigonum, right wall
8	Male	51	Syphilis	Hematuria	265	<40	36	Single, trigonum
9	Male	39	Syphilis	Hematuria	634	6008	24	Single, bladder neck
10	Male	56	Syphilis	Incidental finding	558	NT	36	Single, trigonum
11	Male	28	None	Hematuria	NT	NT	12	Multiple, trigonum
12	Male	67	Hypertension	Hematuria	471	<40	72	Multiple, whole bladder
13	Male	78	Syphilis, hypertension	Hematuria	183	NT	360	Multiple, whole bladder
14	Male	59	Syphilis	Incidental finding	96	<40	60	Multiple, whole bladder
15	Male	58	None	Hematuria	NT	<40	36	Multiple, left, top wall
16	Male	56	Syphilis	Hematuria	460	<40	40	Single, bladder neck
17	Male	60	Hepatitis B	Hematuria	560	<40	36	Multiple, trigonum
18	Male	56	Diabetes	Hematuria	240	<40	36	Single, left wall
19	Male	52	Syphilis, hypertension	Hematuria	681	<40	32	Multiple, right wall
20	Male	56	Diabetes	Incidental finding	360	NT	24	Multiple, left, top wall

Abbreviations: ART: antiretroviral therapy; HIV: human immunodeficiency virus; NT: not tested.

Long-term survival

Median follow-up was 63 months. During this period, seven cancer-specific deaths occurred. Kaplan–Meier analysis yielded 1-, 3-, and 5-years cancer-specific survival (CSS) rates of 85%, 65%, and 60%, respectively (Table 2). On univariate Cox analysis, only tumor stage (stage >T1 vs T1: HR = 17.742; 95% CI: 2.654-148.731; p = 0.006) was significantly associated with CSS. Nadir CD4 count, ART duration, age, and grade were not significant (Table 3).

OPEN IN VIEWER

Table 2.

Cancer-specific survival stratified by clinical and HIV-related variables.

Variables	Patients (n)	Events (n)	Median cancer specific survival (CSS) (months)	1-year CSS	3-years CSS	5-years CSS
Overall	20	7	NR	85%	65%	60%
Age
<60	14	4	NR	88%	62%	62%
>60	6	3	18	75%	50%	NR
Tumour stage
NMIBC	13	2	NR	100%	84%	84%
MIBC	7	5	16	57%	0	0
Tumour grade
Low grade	8	2	NR	100%	88%	NR
High grade	12	5	18	75%	42%	42%
ART duration
<2 years	5	2	NR	100%	60%	60%
>2 years	15	5	NR	80%	59%	59%
CD4+T cell
<200 (cells/ul)	3	1	NR	78%	56%	56%
>200 (cells/ul)	17	6	NR	91%	64%	64%

Abbreviations: CSS: cancer-specific survival; NR: not reached; NMIBC: non-muscle-invasive bladder cancer; MIBC: muscle-invasive bladder cancer.

Events: number of cancer-specific deaths.

OPEN IN VIEWER

Table 3.

Univariable Cox regression analysis for cancer-specific mortality.

Variable	Patients (n)	Cancer specific deaths (n)	HR (95% CI)	p value
Age
<60 years	14	4	Reference	0.161
>60 years	6	3	4.377 (0.866–1.441)
Tumour grade
Low	8	2	Reference	0.230
High	12	5	5.142 (1.451–64.351)
Tumour stage
T1	13	2	Reference	0.006
>T1	7	5	17.742 (2.654–148.731)
Tumour location
Trigone	6	3	Reference	0.070
Other locations	14	4	0.097 (0.008–1.365)
CD4+ T cells
<200 cells/ul	3	1	Reference	0.520
>200 cells/ul	17	6	2.578 (0.445–12.856)
ART duration
<2 years	5	2	Reference	0.757
>2 years	15	5	0.958 (0.097–6.237)

Abbreviations: CI: confidence interval; HR: hazard ratio.

Discussion

The median age at bladder cancer diagnosis in this cohort was 56 years, consistent with prior reports in this population (55–56 years)^8,9 but appeared lower than that in the general population (65–70 years). ¹¹ All patients were male, which aligns with the well-documented male predominance in bladder cancer (3–4 times higher incidence than in females).^5,11,12

In our cohort, 35% of patients had MIBC, of whom 60% had high-grade histology—similar to previous findings in PLWH but appeared higher than in the general population.^8,13 Among NMIBC patients receiving regular intravesical pirarubicin after TURBT, relapse occurred in five (33.3% of those initially diagnosed with stage T1 disease). This relapse rate compared favorably with the >75% recurrence expected without intravesical therapy and the 39% relative risk reduction reported with intravesical chemotherapy. ¹⁴ Only one mild adverse event (bladder irritation) occurred, suggesting pirarubicin is safe in PLWH.

Intravesical bacille Calmette-Guérin (BCG), the preferred treatment for high-risk NMIBC, relies on CD4 + T-cell activation and theoretically risks disseminated infection in PLWH. ¹⁵ While routine use of BCG in this population is not generally recommended, limited evidence—including one patient with HIV who tolerated BCG without complications or relapse ⁹ and three renal transplant recipients who received BCG with prophylactic antituberculosis therapy without systemic infection ¹⁶ —suggests that BCG may be feasible in selected cases with appropriate precautions.

Many studies are underway exploring or have initially confirmed the efficacy and safety of immune checkpoint inhibitors in combination with platinum-based chemotherapy in MIBCs.^17,18 The use of TGC in locally advanced bladder cancer patients was recently demonstrated to have encouraging antitumor activity and good tolerability. ¹⁹ In our study, the confirmed objective responses was 66.7%,which is compared favorably with that reported in recent literature. On the other hand, the incidence of treatment-related adverse events was also similar to that reported previously. ²⁰ Our limited data preliminarily suggest that the TGC regimen is safe and effective, however more definitive conclusions require further validation through larger-sample studies.

Multiple factors influence long-term outcomes in bladder cancer, with tumor stage, grade, gender, and age well-established as prognostic determinants.^21,22 In our cohort, univariate analysis identified tumor stage as the only factor significantly associated with long-term survival; age, tumor grade, and HIV-related parameters (CD4 count, ART duration) were not statistically significant. However, given the small sample size and exploratory nature of these analyses, these findings should be interpreted with caution and require confirmation in larger studies.

Prior studies report 5-years CSS of approximately 90% for males with NMIBC versus 27.1–65.5% for MIBC, depending on substage. ²³ In our series, CSS among NMIBC patients appeared to approximate that of the general population, whereas MIBC patients had 0% 5-years CSS—a marked divergence consistent with findings in other HIV-associated malignancies (e.g., lung, colorectal cancer), where advanced-stage tumors portend worse outcomes.^24,25

This study has several limitations. As a single-center case series, it is subject to inherent biases and lacks a control group of HIV-negative patients, precluding causal inferences about HIV’s impact on bladder cancer outcomes; all comparisons with the general population are descriptive only. The small sample size (n = 20), due to the rarity of this population, limits statistical power and generalizability; our findings are exploratory and require validation in larger, controlled studies. Additionally, smoking data from medical records may be underreported due to incomplete documentation. Despite these limitations, this is the largest single-center series to date on HIV-related bladder cancer, providing valuable descriptive data for future research.

Conclusions

Given the relatively young age and aggressive pathological features observed in this case series, further research is warranted to determine whether targeted screening in PLWH could be beneficial. Limited data preliminarily suggest that intravesical chemotherapy with pirarubicin and adjuvant TGC therapy may be safe and effective options for PLWH with bladder cancer; however, these findings should be considered exploratory and require validation in larger, prospective studies. PLWH with muscle-invasive bladder cancer appear to have poorer CSS, consistent with the prognostic significance of tumor stage observed in the general population.