Background

Adult T-cell leukemia/lymphoma (ATLL) is rare in the United States but clusters in specific racial and geographic sub-populations, presumably mirroring human T-lymphotropic virus-1 (HTLV-1) epidemiology. We re-examined whether household income or county urbanicity modifies risk once age and race are taken into account.

Methods

All malignant ATLL recorded in SEER-17 (2000-2022; N = 699) were linked to county-level median household income and Rural–Urban Continuum codes. Age-, sex- and race-specific person-years were used as offsets in Poisson or negative-binomial models. Cochran–Armitage tests and exact rate ratios explored group differences; a continuous age effect was quantified by Poisson regression.

Results

Incidence rose steeply with age: from 0.4 per 10^6 in individuals < 20 years to 17.0 per 10^6 at ≥ 80 years (trend Z = 26.5, p < 0.0001), equating to a 4.7 % increase in risk per additional year (β = 0.046, p < 0.0001). Race remained the dominant determinant: Black Americans experienced an age-adjusted rate of 10.4 per 10^6, four-fold higher than Whites (2.6 per 10^6) and markedly above Asians (3.6 per 10^6; all p < 0.001). Sex had no influence (3.8 per 10^6 in males vs 3.6 per 10^6 in females; p = 0.45).

In contrast, socio-economic and urban–rural factors showed little impact. Across three income strata (< $60 k, $60–84.9 k, ≥ $85 k), rates were virtually identical at 3.6–3.8 per 10^6 (p = 0.89), and treating income as a continuous variable did not improve model fit. Overall metropolitan and non-metropolitan counties also displayed indistinguishable incidence (3.6 vs 3.7 per 10^6; p = 0.92). However, race-stratified negative-binomial models revealed a metropolitan excess confined to Blacks: after adjusting for age, sex and county income, Black residents of rural counties had a 30 % lower risk than their urban counterparts (Incidence RR non-metro = 0.70, p = 0.01). No urban–rural disparity was detected among Whites or Asians, and adding county income quintiles left this interaction unchanged.

Incidence rose steeply with age, climbing from 0.4 per 10^6 in individuals < 20 y to 17.0 per 10^6 at ≥ 80 y (p < 0.0001). A Poisson model that treated age as a continuous variable yielded β = 0.049 (SE = 0.003), corresponding to a 5.1 % increase in risk per additional year of age (95 % CI 4.3–5.7 %).

Race/ethnicity remained the dominant determinant. After age adjustment, Black Americans showed an incidence of 10.4 per 10^6, four-fold higher than Whites (2.6 per 10^6; RR = 4.0, 95 % CI 3.4–4.7, p<0.0001) and nearly three-fold higher than Asians (3.6 per 10^6; RR = 2.9, 95 % CI 2.2–3.7, p<0.0001). The rate among American Indian/Alaska Native or unknown-race individuals was 3.8 per 10^6 (n = 12) and did not differ significantly from Whites (p = 0.71).

Sex had no measurable influence (3.8 vs 3.6 per 10^6 for males vs females; RR = 1.06, 95 % CI 0.91–1.22, p = 0.48). County-level socio-economic status showed no association: incidence across income tertiles of <$60 k, $60–84.9 k, and ≥$85 k remained virtually identical at 3.6–3.8 per 10^6 (χ² = 0.45, p = 0.80), and modelling income as a continuous predictor did not improve fit (ΔG² = 0.63, p = 0.68).

Overall urbanicity was likewise uninformative—metropolitan and non-metropolitan counties exhibited rates of 3.6 vs 3.7 per 10^6 (χ² = 0.05, p = 0.82). A race-specific interaction, however, was evident: among Blacks, non-metropolitan residence conferred a 30 % lower risk than metropolitan residence (RR = 0.70, 95 % CI 0.53–0.93, p = 0.01); no urban–rural difference was observed in Whites (RR = 0.96, p = 0.66) or Asians (RR = 0.98, p = 0.90).

Conclusions

U.S. ATLL incidence is driven overwhelmingly by age and race, with older Black adults in metropolitan areas carrying the greatest burden. Neither household income nor general urbanicity meaningfully modifies risk outside this subgroup, implying that HTLV-1 exposure patterns and healthcare access—not broad socio-economic status—explain observed disparities. Public-health efforts should therefore target HTLV-1 screening and awareness in middle-aged and older Black urban communities rather than rely on income- or location-based interventions alone.

This content is only available as a PDF.
2025
Sign in via your Institution