About

Lisa Zaba is a Clinical Associate Professor in the Department of Dermatology at Stanford University. She is affiliated with the Center for Artificial Intelligence in Medicine & Imaging (AIMI) at Stanford. Her role involves integrating artificial intelligence into medical imaging and dermatology, contributing to research and education in these fields. Her work focuses on advancing AI applications in healthcare, particularly in dermatology, and she is actively involved in the academic community through her faculty position and participation in AIMI's initiatives.

Research topics

• Medicine
• Pathology
• Dermatology
• Internal medicine
• Intensive care medicine
• Immunology
• Biology
• Genetics

Selected publications

• Response to Gao and Chen, “Comments on Akaike et al’s ‘Circulating tumor DNA level is associated with time to clinical recurrence in Merkel cell carcinoma: Implications for patient management’”

  Journal of the American Academy of Dermatology · 2026-02-17

  articleOpen accessSenior author
  PublisherOA PDFDOI

• Circulating tumor DNA level is associated with time to clinical recurrence in Merkel cell carcinoma: Implications for patient management

  Journal of the American Academy of Dermatology · 2025-10-10 · 2 citations

  articleSenior author
  PublisherDOI

• Development of a Skin-Directed Scoring System for Stevens-Johnson Syndrome and Epidermal Necrolysis

  UNC Libraries · 2025-07-26

  articleOpen access1st authorCorresponding

  Importance: Scoring systems for Stevens-Johnson syndrome and epidermal necrolysis (EN) only estimate patient prognosis and are weighted toward comorbidities and systemic features; morphologic terminology for EN lesions is inconsistent. Objectives: To establish consensus among expert dermatologists on EN terminology, morphologic progression, and most-affected sites, and to build a framework for developing a skin-directed scoring system for EN. Evidence Review: A Delphi consensus using the RAND/UCLA appropriateness criteria was initiated with a core group from the Society of Dermatology Hospitalists to establish agreement on the optimal design for an EN cutaneous scoring instrument, terminology, morphologic traits, and sites of involvement. Findings: In round 1, the 54 participating dermatology hospitalists reached consensus on all 49 statements (30 appropriate, 3 inappropriate, 16 uncertain). In round 2, they agreed on another 15 statements (8 appropriate, 7 uncertain). There was consistent agreement on the need for a skin-specific instrument; on the most-often affected skin sites (head and neck, chest, upper back, ocular mucosa, oral mucosa); and that blanching erythema, dusky erythema, targetoid erythema, vesicles/bullae, desquamation, and erosions comprise the morphologic traits of EN and can be consistently differentiated. Conclusions and Relevance: This consensus exercise confirmed the need for an EN skin-directed scoring system, nomenclature, and differentiation of specific morphologic traits, and identified the sites most affected. It also established a baseline consensus for a standardized EN instrument with consistent terminology.

  PublisherDOI

• Risk of disease progression after discontinuing immunotherapy in 105 patients with Merkel cell carcinoma who responded to PD-1 pathway blockade

  Journal for ImmunoTherapy of Cancer · 2025-08-01 · 2 citations

  articleOpen access

  BACKGROUND: Immune checkpoint inhibitors (ICIs) are the preferred systemic therapy for most patients with advanced Merkel cell carcinoma (MCC). However, the optimal duration of treatment for patients responding to ICI is unclear. Emerging data from retrospective analyses indicate a higher risk of MCC progression after ICI discontinuation, as compared with continuing ICI. METHODS: We performed a retrospective cohort study to evaluate the rate of progressive disease (PD) after treatment discontinuation in patients with advanced MCC who experienced objective responses to first-line ICI. We evaluated whether the risk of PD was associated with the reason for treatment discontinuation (elective vs due to toxicity) and depth of response (complete vs partial response (CR vs PR)). RESULTS: Among 105 responders, 58 discontinued ICI (median treatment duration: 12 months), and 47 continued ICI (median treatment duration: 20 months) at data cut-off. With a median follow-up of 34 months from ICI initiation, 33% of the entire cohort experienced disease progression at 2 years. 2 years after ICI initiation, 39% of patients who discontinued ICI had disease progression, compared with 14% of patients who continued ICI (HR=2.34 (95% CI: 1.07 to 5.12), p=0.034). Among patients who discontinued ICI, those with PR had a numerically higher rate of progression compared with patients with CR at 2 years after ICI discontinuation (56% vs 29%, respectively; HR=1.74 (95% CI: 0.72 to 4.20), p=0.22). Patients who discontinued due to toxicity had numerically higher rates of progression at 2 years (N=28) compared with patients who discontinued electively (N=30) (45% vs 31%, respectively; HR=2.08 (95% CI: 0.79 to 5.46), p=0.14). Among responders who stayed on ICI and had not progressed by 1 year, those who electively discontinued ICI had a high chance of remaining progression-free at 2 years (89%), similar to those who continued ICI (96%, p=0.59). CONCLUSIONS: This study highlights the high progression risk following ICI discontinuation in advanced MCC, especially among patients with non-CRs or those discontinuing early. While elective discontinuation may be appropriate after durable CRs (response≥1 year), greater caution is warranted in other settings.

  PublisherOA PDFDOI

• Dendritic Cells in Normal and Inflamed Human Skin

  Digital Commons - RU (Rockefeller University) · 2025-09-08

  articleOpen access1st authorCorresponding

  Psoriasis is a skin disease originally thought to be a primary keratinocyte differentiation and maturation disease. Several T cell targeted theraputics were found to reverse disease, and thus subsequent research has focussed on the adaptive immune system, particularly effector CD8+ T cells infiltrating the epidermis. Recent studies, however, show that inhibitors of tumor necrosis factor (TNF) are also effective therapeutics. Activated dendritic cells (DCs) produce large ammounts of TNF which acts in an autocrine loop to increase DC maturation. Thus, TNF inhibitors may inhibit DC maturation and downstream T cell activation. This thesis elucidates DC subsets present in normal human skin and psoriasis lesional skin, and the mechanisms by which psoriatic inflammatory DCs activate Th17 T cells and downstream mediators to maintainan psoriatic cutanous inflammation. In normal blood, there exists 3 non-overlapping subsets of myeloid dendritic all of which are Lin-CD11c+HLA-DR+ and either BDCA-1+ (24 ± 2%), CD16+ (70 ± 4%), or BDCA-3+ (5 ± 1%), in order of immunostimulatory capacity. Only two myloid dendritic cell populations exist in normal human dermis: Lin-CD11c+HLA-DR+CD16+ and either BDCA-1+ (≈90%) or BDCA-3+ (≈10%). In situ double-label immunofluorescence showed that approximately 10-15% of CD11c+ dermal cells cluster together in lymphoidlike structures and are BDCA-1+CD205+DC-LAMP+. Upon emigration from the dermis, 90-95% of BDCA-1+ cells expressed these mature antigens, stimulated allogeneic T cells, and increased immunostimulatory capacity after the addition of TNF, PGE2, IL-1β, and IL-6. Functional studies were not performed on BDCA-3+ cells because of limited cell numbers. In normal dermis there also exists a large population of FXIIIA+CD163+ macrophages that are not immunostimulatory and phagocytose large particles. As in normal blood, psoriasis patient blood contained 3 non-overlapping subsets of myeloid DCs (BDCA-1+, CD16+, or BDCA-3+). In psoriatic skin the frequency and distribution of BDCA-1+ and BDCA-3+ cells is similar to normal, however, there was a 30-fold increase in "inflammatory" CD11c+ cells that did not express either marker. Most BDCA-1+ cells expressed maturation markers CD205 and DC-LAMP, while most BDCA-1- inflammatory cells expressed CD209 immature DC/ macrophage marker. Some myeloid cells expressed TNF and inducible nitric oxide synthase (iNOS). Treatment of psoriasis patients with the TNF neutralizing antibody etanercept not only inhibited dendritic cells as expected, but also had inhibitory effects on a newly appreciated type of T cell – Th17 cells. Etanercept reduced inflammatory DC products that drive Th17 cell proliferation (IL-23) as well as Th17 products and downstream effector mollecules (IL-17, IL-22, CCL20, and DEFB4). In contrast, Th1 cellular products and effector molecules (IFNγ, LTA-1, and MX-1) were reduced late in disease resolution. Using affymetrix gene array we characterized a global set of 4 gene clusters modulated temporally over the course of etanercept treatment. TNF and IL-17 pathway genes were downmodulated with a similar velocity, while IFNγ pathway genes were downmodulated later.

  PublisherOA PDFDOI

• Comparison of surveillance circulating tumor DNA and Merkel polyomavirus antibody titer for detection of Merkel cell carcinoma recurrence.

  Journal of Clinical Oncology · 2025-05-28

  articleSenior author

  9574 Background: Circulating tumor DNA (ctDNA) is emerging as a robust biomarker for detecting recurrences in Merkel cell carcinoma (MCC). This study aims to compare the performance of ctDNA against the Merkel polyomavirus antibody titer (AMERK) test in predicting MCC recurrence risk. Methods: We conducted a longitudinal, multi-center observational study involving 171 MCC patients undergoing disease surveillance, including serial ctDNA and AMERK testing (median testing interval: 92 days). All patients had detectable antibodies by AMERK at initial diagnosis and both tests were conducted within 45 days of each other. ctDNA tests were classified as positive if ctDNA was > 0 MTM/mL. An AMERK test was positive if antibody titers rose ≥ 30% from the prior titer. Clinical recurrences were identified through routine imaging and clinical examinations. The diagnostic performance of ctDNA and AMERK tests was assessed using positive and negative predictive values (PPV and NPV), recurrence-free survival after any positive test vs. all negative tests, and corresponding hazard ratios (HRs) from Cox regression. Results: 718 pairs of ctDNA and AMERK tests were collected from 171 patients. Over a median follow-up of 445 days, there were 38 clinical recurrences, 91/718 (13%) positive ctDNA tests, and 73/718 (10%) positive AMERK tests. A significantly increased clinical recurrence rate was observed in patients with a positive ctDNA test compared to those with consistently negative results (HR: 27.4, 95%CI: 11.0-68.3) (Table 1). Although a positive AMERK test was similarly associated with higher clinical recurrence (HR: 5.8, 95%CI: 3.0-11.1), the rate was distinctly lower than that for a positive ctDNA test (HR: 5.8 vs. 27.4; p < 0.001). The PPV for clinical recurrence at 1 year after a positive test was significantly higher for ctDNA vs. AMERK (PPV: 73% [95% CI: 58-84%] vs. 51% [95% CI: 29-70%]; p=0.014). NPV for recurrence within 4 months of a negative test for the ctDNA test was similarly higher for ctDNA vs. AMERK (NPV: 98% [95% CI: 97-99%] vs. 95% [95% CI: 92-97%]; p=0.001). The median lead time between the first positive test and a clinically detected recurrence was 3.1 months for ctDNA (among 30 recurrences preceded by a positive test) and 1.9 months for AMERK (among 19 recurrences preceded by a positive test) (p=0.063). Conclusions: Our results indicate that, in a cohort of AMERK positive patients, ctDNA outperforms AMERK for detection of MCC recurrence. ctDNA may be a viable alternative to AMERK in clinical practice and may better identify high-risk patients who benefit from more aggressive monitoring or adjuvant therapy trials. Hazard ratios for subsequent MCC clinical recurrence: Comparison of positive ctDNA and AMERK tests. Test HR (95% CI) P-value Positive ctDNA 27.4 (11.0, 68.3) <0.001 Positive AMERK 5.8 (3.0, 11.1) <0.001 Difference (ctDNA / AMERK) 4.7 (2.1, 15.9) <0.001

  PublisherDOI

• Risk factors for regional or distant metastatic disease in sebaceous carcinoma: A retrospective cohort study from the National Cancer Data Base

  Journal of the American Academy of Dermatology · 2025-09-11

  article
  PublisherDOI

• 0349 A longitudinal study of stress and depression in patients with hidradenitis suppurativa

  Journal of Investigative Dermatology · 2025-07-21

  articleOpen access
  PublisherOA PDFDOI

• Automated Detection of Benign and Malignant Skin Lesions from Reflectance Confocal Microscopy Images Using Deep Learning

  JID Innovations · 2025-08-08 · 2 citations

  articleOpen accessSenior author

  Reflectance confocal microscopy offers a noninvasive approach for diagnosing skin lesions at the point of care, but it remains underutilized owing to the specialized skill required for interpretation. Artificial intelligence provides an opportunity to automate this process. We developed deep learning models to automate the analysis of reflectance confocal microscopy block images. Reflectance confocal microscopy images acquired from 3rd and 4th generation VivaScope 1500 devices were preprocessed and split for training and testing. Two models were developed: a modified convolutional neural network ResNet-18, for skin layer detection, and a ResNet-34 integrated with a gated recurrent unit for lesion classification. The models were pretrained on 3rd generation images and fine tuned on 4th generation data, utilizing 5-fold cross-validation. Our cohort included 845 patients, 1147 lesions, and 4391 VivaBlock images. The layer detection model identified the dermis, epidermis, and dermoepidermal junction, achieving an area under the curve of 0.70, 0.71, and 0.57, respectively. The lesion classification model distinguished malignant from benign lesions with an area under the curve of 0.80 and specificity of 0.91. Our convolutional neural network gated recurrent unit approach effectively distinguished benign from malignant lesions, showing impressive diagnostic accuracy mimicking expert dermatological assessments. This highlights artificial intelligence's potential in improving reflectance confocal microscopy image interpretation, reducing unnecessary biopsies, and paves the way for future research.

  PublisherDOI

• 232 How do flares impact psychosocial outcomes in patients with Hidradenitis Suppurativa (HS)?

  Journal of Investigative Dermatology · 2025-11-01

  article
  PublisherDOI

Frequent coauthors

• Paul Nghiem

  University of Washington

  39 shared

• Daniel S. Hippe

  Fred Hutch Cancer Center

  31 shared

• Nolan J. Maloney

  Stanford University

  28 shared

• Thomas H. Pulliam

  27 shared

• Kelly G. Paulson

  University of Washington Medical Center

  24 shared

• Xinyi Fan

  Fred Hutch Cancer Center

  23 shared

• Ata S. Moshiri

  NYU Langone Health

  23 shared

• Rima M. Kulikauskas

  23 shared