Comparison of humans versus mobile phone-powered artificial intelligence for the diagnosis and management of pigmented skin cancer in secondary care: a multicentre, prospective, diagnostic, clinical trial.

BACKGROUND: Diagnosis of skin cancer requires medical expertise, which is scarce. Mobile phone-powered artificial intelligence (AI) could aid diagnosis, but it is unclear how this technology performs in a clinical scenario. Our primary aim was to test in the clinic whether there was equivalence between AI algorithms and clinicians for the diagnosis and management of pigmented skin lesions. METHODS: In this multicentre, prospective, diagnostic, clinical trial, we included specialist and novice clinicians and patients from two tertiary referral centres in Australia and Austria. Specialists had a specialist medical qualification related to diagnosing and managing pigmented skin lesions, whereas novices were dermatology junior doctors or registrars in trainee positions who had experience in examining and managing these lesions. Eligible patients were aged 18-99 years and had a modified Fitzpatrick I-III skin type; those in the diagnostic trial were undergoing routine excision or biopsy of one or more suspicious pigmented skin lesions bigger than 3 mm in the longest diameter, and those in the management trial had baseline total-body photographs taken within 1-4 years. We used two mobile phone-powered AI instruments incorporating a simple optical attachment: a new 7-class AI algorithm and the International Skin Imaging Collaboration (ISIC) AI algorithm, which was previously tested in a large online reader study. The reference standard for excised lesions in the diagnostic trial was histopathological examination; in the management trial, the reference standard was a descending hierarchy based on histopathological examination, comparison of baseline total-body photographs, digital monitoring, and telediagnosis. The main outcome of this study was to compare the accuracy of expert and novice diagnostic and management decisions with the two AI instruments. Possible decisions in the management trial were dismissal, biopsy, or 3-month monitoring. Decisions to monitor were considered equivalent to dismissal (scenario A) or biopsy of malignant lesions (scenario B). The trial was registered at the Australian New Zealand Clinical Trials Registry ACTRN12620000695909 (Universal trial number U1111-1251-8995). FINDINGS: The diagnostic study included 172 suspicious pigmented lesions (84 malignant) from 124 patients and the management study included 5696 pigmented lesions (18 malignant) from the whole body of 66 high-risk patients. The diagnoses of the 7-class AI algorithm were equivalent to the specialists' diagnoses (absolute accuracy difference 1·2% [95% CI -6·9 to 9·2]) and significantly superior to the novices' ones (21·5% [13·1 to 30·0]). The diagnoses of the ISIC AI algorithm were significantly inferior to the specialists' diagnoses (-11·6% [-20·3 to -3·0]) but significantly superior to the novices' ones (8·7% [-0·5 to 18·0]). The best 7-class management AI was significantly inferior to specialists' management (absolute accuracy difference in correct management decision -0·5% [95% CI -0·7 to -0·2] in scenario A and -0·4% [-0·8 to -0·05] in scenario B). Compared with the novices' management, the 7-class management AI was significantly inferior (-0·4% [-0·6 to -0·2]) in scenario A but significantly superior (0·4% [0·0 to 0·9]) in scenario B. INTERPRETATION: The mobile phone-powered AI technology is simple, practical, and accurate for the diagnosis of suspicious pigmented skin cancer in patients presenting to a specialist setting, although its usage for management decisions requires more careful execution. An AI algorithm that was superior in experimental studies was significantly inferior to specialists in a real-world scenario, suggesting that caution is needed when extrapolating results of experimental studies to clinical practice. FUNDING: MetaOptima Technology.

Clinical Outcomes of 2-Drug Regimens vs 3-Drug Regimens in Antiretroviral Treatment-Experienced People Living With Human Immunodeficiency Virus.

BACKGROUND: Limited data exist that compare clinical outcomes of 2-drug regimens (2DRs) and 3-drug regimens (3DRs) in people living with human immunodeficiency virus. METHODS: Antiretroviral treatment-experienced individuals in the International Cohort Consortium of Infectious Diseases (RESPOND) who switched to a new 2DR or 3DR from 1 January 2012-1 October 2018 were included. The incidence of clinical events (AIDS, non-AIDS cancer, cardiovascular disease, end-stage liver and renal disease, death) was compared between regimens using Poisson regression. RESULTS: Of 9791 individuals included, 1088 (11.1%) started 2DRs and 8703 (88.9%) started 3DRs. The most common 2DRs were dolutegravir plus lamivudine (22.8%) and raltegravir plus boosted darunavir (19.8%); the most common 3DR was dolutegravir plus 2 nucleoside reverse transcriptase inhibitors (46.9%). Individuals on 2DRs were older (median, 52.6 years [interquartile range, 46.7-59.0] vs 47.7 [39.7-54.3]), and a higher proportion had ≥1 comorbidity (81.6% vs 73.9%). There were 619 events during 27 159 person-years of follow-up (PYFU): 540 (incidence rate [IR] 22.5/1000 PYFU; 95% confidence interval [CI]: 20.7-24.5) on 3DRs and 79 (30.9/1000 PYFU; 95% CI: 24.8-38.5) on 2DRs. The most common events were death (7.5/1000 PYFU; 95% CI: 6.5-8.6) and non-AIDS cancer (5.8/1000 PYFU; 95% CI: 4.9-6.8). After adjustment for baseline demographic and clinical characteristics, there was a similar incidence of events on both regimen types (2DRs vs 3DRs IR ratio, 0.92; 95% CI: .72-1.19; P = .53). CONCLUSIONS: This is the first large, international cohort to assess clinical outcomes on 2DRs. After accounting for baseline characteristics, there was a similar incidence of events on 2DRs and 3DRs. 2DRs appear to be a viable treatment option with regard to clinical outcomes. Further research on resistance barriers and long-term durability of 2DRs is needed.

Prevalence of Skin-specific Autoantibodies in HIV-infected Patients and Uninfected Controls.

Various autoantibodies are detected more frequently in HIV-infected individuals than in HIV-negative controls; however, limited data exist regarding autoimmune blistering skin diseases. Using enzyme-linked immunoassay (ELISA) and indirect immunofluore-scence, no difference in the frequency and magnitude of autoantibodies against BP180, BP230, desmoglein 1 and 3 was found between 594 HIV-infected patients and 248 uninfected controls in this cross-sectional study (16.0% vs. 11.7%, respectively, for at least one positive ELISA, p = 0.11). Interestingly, reactive syphilis serology in both HIV-infected individuals and uninfected controls was associated with positive anti-BP180 ELISA results (adjusted odds ratio (OR) 2.14, 95% confidence interval (CI) 1.07-4.29, p = 0.03 and OR 4.70, CI 1.3-16.86; p = 0.0180). Our study shows a comparably low prevalence of cutaneous autoantibodies in both HIV-infected patients and uninfected controls lacking signs of autoimmune blistering skin disease. Positive BP180 ELISA in the absence of clinical signs of bullous pemphigoid should prompt further evaluation for syphilis antibodies.

Acute HIV Infection Results in Subclinical Inflammatory Cardiomyopathy.

The impact of excess viral RNA on myocardial function and morphology in the setting of acute human immunodeficiency virus (HIV) infection remains unknown. In this study, 49 patients with acute HIV infection showed increased levels of N-terminal prohormone of brain natriuretic peptide, a surrogate of myocardial function, which decreased with viral suppression and normalization of systemic inflammation (79 pg/mL vs 28 pg/mL; P < .001). A comparable change was seen with levels of troponin T, a marker of morphologic myocardial damage (4.9 ng/L vs 1.5 ng/L; P < .001). In conclusion, we observed significant functional and morphological myocardial impairment during acute HIV infection, fueled by inflammatory activation and extensive viral replication, resulting in a reversible subclinical inflammatory cardiomyopathy.

Epidemiology of Human Herpes Virus 8 in Pregnant Women and their Newborns--A cross-sectional delivery survey in Central Gabon.

OBJECTIVES: On the background of a high prevalence of HHV-8 infection in pre-pubertal Central African children, this study investigated the potential for in utero transmission of HHV-8. PATIENTS: Gabonese pregnant women were invited to provide peripheral and cord blood samples for serological and PCR diagnostics of HHV-8 infection at delivery for this cross-sectional survey. RESULTS: Out of 344 participants 120 (35%, 95% CI: 30-40%) were serologically positive for HHV-8. 31% (95% CI: 22-40%) of cord blood samples of seropositive women had detectable IgG antibodies. Among all seropositive participants HHV-8 was detected by PCR in one maternal peripheral blood sample at delivery (1%, 95% CI: 0.2-7%) and in none of cord blood samples. There was no association between demographic characteristics and infection status. Similarly, there was no difference in risk for premature delivery, low birth weight, and maternal anaemia in HHV-8 seropositive women. DISCUSSION: These data suggest a high seroprevalence of HHV-8 infection in pregnant women, however viraemia at delivery does not commonly occur in Central Africa. Based on these observations it may be speculated that infection of children may occur more commonly either antepartum or later on in infancy and childhood.