Fred M. Baik

Background: The clinical relevance of antibiotic exposure on immune checkpoint inhibitor (ICI) efficacy in head and neck squamous cell carcinoma (HNSCC) is not well characterized. Methods: This retrospective study included patients with recurrent/metastatic HNSCC treated between 2015 and 2025. Cox proportional hazards and Kaplan–Meier method modeled overall survival (OS). Results: Of 316 patients (median age 68 years, 70.9% male), primary tumor site was 40.5% oral cavity, 34.8% oropharynx, 14.9% sinonasal/nasopharynx, 9.8% hypopharynx/larynx. One hundred and forty‐eight (46.8%) patients received antibiotics surrounding ICI. Antibiotics were associated with decreased OS (HR = 1.53, 95% CI 1.15–2.04). On multivariable analysis, while intravenous route of administration and spectrum of antibiotic activity were not risk factors, oral cavity site (HR = 1.72, 95% CI 1.07–2.76), ECOG 2–4 (HR = 2.78, 95% CI 1.70–3.67), PD‐L1 CPS < 1 (HR = 3.16, 95% CI 1.79–5.04), and number of antibiotic classes (HR = 1.37, 95% CI 1.18–1.59) independently predicted decreased OS. Conclusions: Antibiotic exposure was associated with decreased survival in HNSCC treated with ICI; however, future studies are needed to evaluate potential mechanistic etiologies underlying this association.

Background: Elective neck dissection (END) is the standard treatment for early‐stage oral cavity squamous cell carcinoma (OCSCC). This study compared costs of sentinel lymph node biopsy (SLNB) and END from the Medicare payer perspective. Methods: Thirty‐two patients with T1‐2N0M0 OCSCC (16 SLNB and 16 END) treated at an academic institution from 2017 to 2025 were included. Results: 32 patients were included: median age 62, 46.9% male, 38% had adjuvant RT. The SLNB cohort had shorter operative time and length‐of‐stay. Four patients had reoperations: 2 after SLNB (completion neck dissection, tumor re‐resection) and 2 after END (neck hematomas). The SLNB cohort had lower costs for postoperative admission ($9567 vs. $12 920, p = 0.015). Total costs were not statistically different ($19 777 vs. $22 126, p = 0.45). On break‐even analysis, SLNB with a completion ND rate ≤ 35% had lower costs than END. Conclusions: Costs of SLNB were 10.6% less than END, and remained lower with completion ND rates ≤ 35%.

We investigate how vibrotactile wrist feedback can enhance spatial guidance for handheld tool movement in optical see-through augmented reality (AR). While AR overlays are widely used to support surgical tasks, visual occlusion, lighting conditions, and interface ambiguity can compromise precision and confidence. To address these challenges, we designed a multimodal system combining AR visuals with a custom wrist-worn haptic device delivering directional and state-based cues. A formative study with experienced surgeons and residents identified key tool maneuvers and preferences for reference mappings, guiding our cue design. In a cue identification experiment (N=21), participants accurately recognized five vibration patterns under visual load, with higher recognition for full-actuator states than spatial direction cues. In a guidance task (N=27), participants using both AR and haptics achieved significantly higher spatial precision (5.8 mm) and usability (SUS = 88.1) than those using either modality alone, despite having modest increases in task time. Participants reported that haptic cues provided reassuring confirmation and reduced cognitive effort during alignment. Our results highlight the promise of integrating wrist-based haptics into AR systems for high-precision, visually complex tasks such as surgical guidance. We discuss design implications for multimodal interfaces supporting confident, efficient tool manipulation.

Purpose: In this paper, we develop and clinically evaluate a depth-only, markerless augmented reality (AR) registration pipeline on a head-mounted display, and assess accuracy across small or low-curvature anatomies in real-life operative settings. Methods: On HoloLens 2, we align Articulated HAnd Tracking (AHAT) depth to Computed Tomography (CT)-derived skin meshes via (i) depth-bias correction, (ii) brief human-in-the-loop initialization, (iii) global and local registration. We validated the surface-tracing error metric by comparing "skin-to-bone" relative distances to CT ground truth on leg and foot models, using an AR-tracked tool. We then performed seven intraoperative target trials (feet x2, ear x3, leg x2) during the initial stage of fibula free-flap harvest and mandibular reconstruction surgery, and collected 500+ data per trial. Results: Preclinical validation showed tight agreement between AR-traced and CT distances (leg: median |Delta d| 0.78 mm, RMSE 0.97 mm; feet: 0.80 mm, 1.20 mm). Clinically, per-point error had a median of 3.9 mm. Median errors by anatomy were 3.2 mm (feet), 4.3 mm (ear), and 5.3 mm (lower leg), with 5 mm coverage 92-95%, 84-90%, and 72-86%, respectively. Feet vs. lower leg differed significantly (Delta median ~1.1 mm; p < 0.001). Conclusion: A depth-only, markerless AR pipeline on HMDs achieved ~3-4 mm median error across feet, ear, and lower leg in live surgical settings without fiducials, approaching typical clinical error thresholds for moderate-risk tasks. Human-guided initialization plus global-to-local registration enabled accurate alignment on small or low-curvature targets, improving the clinical readiness of markerless AR guidance.

Optical see-through augmented reality (OST-AR) systems like Microsoft HoloLens 2 hold promise for arm's distance guidance (e.g., surgery), but depth perception of the hologram and occlusion of real instruments remain challenging. We present an evaluation of how visualizing the target object with different transparencies and visualizing a tracked tool (virtual proxy vs. real tool vs. no tool tracking) affects depth perception and system usability. Ten participants performed two experiments on HoloLens 2. In Experiment 1, we compared high-transparency vs. low-transparency target rendering in a depth matching task at arm's length. In Experiment 2, participants performed a simulated surgical pinpoint task on a frontal bone target under six visualization conditions ($2 \times 3$: two target transparencies and three tool visualization modes: virtual tool hologram, real tool, or no tool tracking). We collected data on depth matching error, target localization error, system usability, task workload, and qualitative feedback. Results show that a more opaque target yields significantly lower depth estimation error than a highly transparent target at arm's distance. Moreover, showing the real tool (occluding the virtual target) led to the highest accuracy and usability with the lowest workload, while not tracking the tool yielded the worst performance and user ratings. However, making the target highly transparent, while allowing the real tool to remain visible, slightly impaired depth cues and did not improve usability. Our findings underscore that correct occlusion cues, rendering virtual content opaque and occluding it with real tools in real time, are critical for depth perception and precision in OST-AR. Designers of arm-distance AR systems should prioritize robust tool tracking and occlusion handling; if unavailable, cautiously use transparency to balance depth perception and tool visibility.

Purpose Sentinel lymph node biopsy (SLNB) helps stage melanoma. Pre-surgical single-photon emission computed tomography/computed tomography (SPECT/CT) visualizes draining lymph nodes, but intraoperative gamma probe detection only estimates SLN location. This study evaluates augmented reality (AR) for projecting pre-surgical SLN imaging onto patients to aid precise localization and extraction. Methods Molecular sieves (8 mm) incubated in fluorine-18 simulated lymph nodes and were implanted in the head and neck region of cadavers. Positron emission tomography/magnetic resonance imaging (PET/MRI) replaced SPECT/CT due to institutional restriction on cadavers. Virtual PET/MRI renderings were projected using the HoloLens 2 and custom software. Five cadavers underwent surgeries with standard, AR, and AR with head movement compensation methods. Results AR achieved a mean surface localization error of 2.5±2.0 mm (range, 0–8 mm) and a depth error of 2.3±1.7 mm (range, 1–7 mm), both within PET voxel resolution. For more challenging level V nodes, the mean surface error slightly increased to 2.9 mm. Compared to manual surface marking, which had an average error of 18.6±13.0 mm (range, 6–62 mm), the AR system significantly reduced errors both in the head-straight and rotated positions (p <.001). Additionally, the AR system reduced the task completion time by 74% (35.1/47.4 s), with an average time of 12.3 s compared to 47.4 s for manual methods. Conclusion The AR system demonstrated high accuracy and efficiency in SLN localization, integrating head-movement compensation and 3D visualization to improve precision and reduce operating room time. Clinical trial number Not applicable. Graphic abstract

Background: Sentinel lymph node biopsy (SLNB) is integral to management of head and neck melanomas (HNM). Here, we evaluate how radiotracer selection (Tc-99m tilmanocept or Tc-99m sulfur colloid) and imaging modality (planar imaging or single positron emission computed tomography (SPECT)) may influence SLNB and its outcomes. Methods: We performed a retrospective chart review of patients who received SLNB for HNM at our institution between January 2000 and July 2022. Results: A total of 211 patients received Tc-99m tilmanocept (Tc-tilmanocept) and 113 received Tc-99m sulfur colloid (Tc-SC). There was no difference between Tc-tilmanocept and Tc-SC in terms of number of nodal levels to which the radiotracers localized (1.48 ± 0.93 SD vs 1.38 ± 0.93, respectively). Tc-tilmanocept had a lower rate of failed sentinel node identification as reported on radiology reports compared to Tc-SC (1.9% vs 11.5%, P < .001). There was a lower 5-year recurrence rate with Tc-tilmanocept versus Tc-SC (20.9% vs 28.3%, respectively P = .07), but this difference was not significant. When analyzing only cases in which SPECT was obtained, this there was no observed difference in 5-year recurrence rate. With SPECT, fewer patients had non-specific description of their node localization as “cervical” without further detailing of nodal basin (0% SPECT vs 9.7% non-SPECT, P < .001). Multivariable linear regression analysis for predictors of HNM 5-year recurrence found the odds ratio for SPECT imaging was 0.43, but not statistically significant ( P = .06). Conclusions: Tc-tilmanocept may be associated with greater rate of successful radiotracer migration along lymphatics compared to Tc-SC. SPECT imaging may confer more precision of the nodal basin to which radiotracer localizes as fewer patients receiving SPECT had radiology findings showing nonspecific localization of nodes to “cervical” basins without further specification. More patients are necessary to determine if Tc-tilmanocept and SPECT imaging are associated with lower melanoma recurrence rate.

PURPOSE: To evaluate the feasibility of administering three doses of neoadjuvant atezolizumab before curative surgical resection in patients with advanced cutaneous squamous cell carcinoma (cSCC). PATIENTS AND METHODS: This single-arm phase II trial included patients with cSCC of the head and neck with stage III or IV disease, or stage II lesions for which standard therapy would incur an unacceptable morbidity. Patients received up to three doses of atezolizumab at a fixed dose of 1,200 mg every 21 days before undergoing surgical resection. The primary end point was the percentage of patients able to complete three doses of therapy and be eligible for surgical resection without discontinuation due to toxicity or progression. Secondary end points included unconfirmed RECIST 1.1 response rate, pathological response rate (major and complete pathological response), and safety and tolerability of atezolizumab. RESULTS: Twenty patients were enrolled and treated. Sixteen (80%) of 20 patients completed three doses of atezolizumab, and all patients were eligible for surgical resection. A pathological complete response was seen in seven (35%; 95% CI, 15.4 to 59.2) patients, and a major pathological response (<10% viable tumor) was seen in four (20%; 95% CI, 5.7 to 43.7) patients. RECIST responses were seen in eight (40%; 95% CI, 19.1 to 64.0) patients, and one (5%) patient progressed. Grade 3 or higher adverse events were seen in one (5%) patient who developed pneumonitis after the first dose of atezolizumab. CONCLUSION: Neoadjuvant atezolizumab is feasible and well tolerated in patients with advanced cSCC and results in a high rate of major and complete pathological responses.