 |
|
|
| |
Primary CNS Lymphoma Treatment Less Aggressive With HIV, But Survival Similar
|
| |
| |
IDWeek 2018, October 3-7, 2018, San Francisco
Mark Mascolini
People with HIV infection and primary central nervous system lymphoma (PCNSL) differ from HIV-negative PCNSL patients in imaging results and treatment, according to a 144-person analysis [1]. Much lower proportions of HIV-positive patients received chemotherapy or autologous stem cell transplant. But HIV infection did not predict faster PCNSL progression or shorter survival.
PCNSL is a rare form of non-Hodgkin lymphoma, an AIDS-defining cancer. For HIV-positive people with PCNSL, clinicians adapted management strategies used in the general population. But the validity of those strategies in people with HIV is poorly understood because HIV patients usually get excluded from prospective studies of PCNSL. In the current antiretroviral era, little is known about potential differences in PCNSL clinical characteristics, prognosis, or treatment strategies or outcomes in people with versus without HIV. To address those questions, Baylor College of Medicine researchers conducted this retrospective study.
The analysis involved everyone diagnosed with PCNSL at Houston's MD Anderson Cancer Center in 2000-2017 or at Ben Taub Hospital in 2012-2016. Participants had no evidence of systemic lymphoma on whole-body CT or PET scan or by bone marrow biopsy. Of the 144 patients with PCNSL, 27 (19%) had HIV infection. At PCNSL diagnosis, only 8 people with HIV (30%) were taking antiretroviral therapy, 74% had a detectable viral load, and 85% had a CD4 count below 200.
Compared with HIV-negative PCNSL patients, those with HIV were younger at diagnosis (median 38 versus 63, P = 0.001), more likely to be men (74% versus 49%, P = 0.02), and more likely to be black (59% versus 7%, P = 0.001). MRI findings indicated multiple lesions more often in the HIV group (67% versus 43%, P = 0.02) and more frequent hemorrhage (59% versus 37%, P = 0.03). Tumor location and size did not differ between people with and without HIV. Presenting symptoms did not differ significantly by HIV status, but people with HIV had significantly lower performance status and higher (worse) ECOG scores.
Initial treatment rates differed substantially by HIV status. People with HIV were most likely to receive initial whole-brain radiation therapy (63%), followed by chemotherapy (22%) and supportive care (4%). In contrast, 95% of HIV-negative people received initial chemotherapy, 15% got supportive care, and 3% received whole-brain radiation therapy. Twenty-four patients--all of them without HIV--underwent autologous stem-cell transplant.
Survival differed markedly by type of treatment. Median overall survival measured only 1 month in untreated people, 24 months for whole-brain radiation therapy alone, and 47 months in people receiving chemotherapy (P < 0.01).
In multivariate analysis, not receiving stem-cell transplant tripled the risk of death (hazard ratio [HR] 3, 95% confidence interval [CI] 1 to 11, P = 0.04). Not undergoing stem-cell transplant emerged as the only independent predictor of disease progression (HR 5, 95% CI 1 to 17, P = 0.01). Compared with no treatment, risk of death was similarly low with chemotherapy (HR 0.05, 95% CI 0 to 0.5, P = 0.01) and whole-brain radiation therapy (HR 0.06, 95% CI 0 to 0.7, P = 0.02). Multivariate analysis did not link HIV infection to shorter overall survival or progression-free survival.
The Baylor investigators concluded that PCNSL imaging results and treatment strategies differ between people with and without HIV infection. Getting methotrexate-based chemotherapy and undergoing autologous stem-cell transplant strongly predicted better outcomes--and PCNSL patients without HIV received these treatments much more often than people with HIV. Yet HIV did not independently predict mortality or PCNSL progression in multivariate analysis.
Reference
1. Dandachi D, Ostrom Q, Insun Chong I, Bondy M, Jose Serpa J, Rivka Colen R, Moron F. Primary central nervous system lymphoma in patients with HIV and non-HIV: should we treat them differently? IDWeek 2018, October 3-7, 2018, San Francisco. Abstract 2245. Poster at https://idsa.confex.com/idsa/2018/webprogram/Paper72325.html
program abstract
Background:
Primary central nervous system lymphoma (PCNSL) is a rare type of non-Hodgkin lymphoma, mostly diffuse large B-cell type. In patients living with HIV (PLWH), PCNSL is associated with Epstein-Barr virus. The optimal diagnostic and prognostic tools, and treatment are yet to be defined. PLWH are typically excluded from prospective studies. The management of PCNSL is adopted from immunocompetent patients.
Methods:
We retrospectively reviewed 122 PCNSL cases presenting to MD Anderson Cancer Center from 2000-2016 (n=84) and Ben-Taub Hospital from 2012-2016 (n=38) to evaluate and compare the clinical characteristics, management, and clinical outcomes in patients with or without HIV infection.
Results:
Among 122 PCNSL cases, 21% had positive HIV test, of those, 89% had CD4 < 200 and 77% were not on antiretrovirals and not virally suppressed. PLWH were significantly younger (37 vs. 62 yrs. P<0.01), and more likely to be African-Americans (61% vs. 7%; p<0.01) and males (73% vs. 50%; p=0.04) than non-HIV patients. There were no differences in presenting symptoms, ocular involvement, B-symptoms, and deep brain involvement. PLWH were more likely to have multiple brain lesions (69% vs. 44%, p=0.02). Immunohistochemistry prognostic markers and the International Extranodal Lymphoma Study Group (IELSG) prognostic score were not different between HIV and non-HIV patients. Nevertheless, treatment strategies varied significantly. PLWH were more likely to receive whole brain radiation therapy as sole treatment (65% vs. 4%) and palliative care (12% vs. 2%), and less likely to receive chemotherapy (23% vs. 94%) (p<0.01). Also, 13% of the patients (all non-HIV) underwent autologous stem cell transplant. Most PLWH (88%) started antiretroviral therapy after diagnosis. Higher IELSG score was an independent predictor of mortality in multivariate regression analysis. The 2-year survival did not differ between PLWH and non-HIV patients [46% (30% - 72%) versus 61% (52% - 72%) (p=0.12)].
Conclusion:
Variation in the treatment of PCNSL between HIV and non-HIV patients is not fully explained by baseline characteristics and prognostic factors. More efforts are needed to identify causes underlying these disparities and ways to alleviate them.
|
| |
|
|
|
|
|
