No abstract

BackgroundImmune checkpoint inhibiting antibodies were introduced into routine clinical practice for cancer patients. Checkpoint blockade has led to durable remissions in some patients, but may also induce immune-related adverse events (irAEs). Lung cancer patients show an increased risk for complications, when infected with influenza viruses. Therefore, vaccination is recommended. However, the efficacy and safety of influenza vaccination during checkpoint blockade and its influence on irAEs is unclear. Similarly, the influence of vaccinations on T cell-mediated immune reactions in patients during PD-1 blockade remains poorly defined.MethodsWe vaccinated 23 lung cancer patients and 11 age-matched healthy controls using a trivalent inactivated influenza vaccine to investigate vaccine-induced immunity and safety during checkpoint blockade.ResultsWe did not observe significant differences between patients and healthy controls in vaccine-induced antibody titers against all three viral antigens. Influenza vaccination resulted in protective titers in more than 60% of patients/participants. In cancer patients, the post-vaccine frequency of irAEs was 52.2% with a median time to occurrence of 3.2 months after vaccination. Six of 23 patients (26.1%) showed severe grade 3/4 irAEs. This frequency of irAEs might be higher than the rate previously published in the literature and the rate observed in a non-study population at our institution (all grades 25.5%, grade 3/4 9.8%).ConclusionsAlthough this is a non-randomized trial with a limited number of patients, the increased rate of immunological toxicity is concerning. This finding should be studied in a larger patient population.

The purpose of this study was to evaluate the expression of the Ca(2+)-binding S100 proteins S100A1, S100A2, S100A3, S100A4, S100A6 and S100B in normal skin. These immunohistochemical staining patterns were compared with those in melanocytic lesions. Paraffin-embedded tissue of normal skin adjacent to 26 naevi, 39 primary cutaneous melanomas and 14 cutaneous melanoma metastases was incubated with polyclonal antibodies against the recombinant human S100 proteins (S100A1, S100A2, S100A3, S100A4, S100A6, S100B) using the alkaline phosphatase anti-alkaline phosphatase method. The S100A2 antibody stained the basal layer of the epidermis and hair follicles of normal skin. Four of 39 primary cutaneous melanomas were positive for S100A2, whereas none of the metastases or naevi showed any immunoreactivity. The S100A3 antibody only stained the inner root sheath cuticle of some hair follicles but no melanocytes or melanocytic lesions. Staining of S100A4 was weak and thus omitted to further analysis. S100A6 faintly labelled keratinocytes. Langerhans' cells, melanocytes and sweat glands. S100A6 immunoreaction was found in two of seven junctional naevi, five of seven compound naevi, and all dermal and blue naevi. There was an intense cytoplasmatic reaction for S100A6 in all primary cutaneous melanomas and in nine of 14 (64%) metastases, S100B was positive in melanocytes and Langerhans' cells, all primaries as well as in the metastases, S100A1 protein was not detected on any of the tissue specimens examined. Whereas S100B and S100A6 antibodies are useful markers for malignant melanoma, expression of S100A4 antibody is too low to be used for immunohistochemical staining. S100A1 and S100A3 antibodies are not expressed in melanocytic lesions and S100A2 is only found in selected tumours. The investigated S100 proteins, including S100B and S100A6, are also expressed in selected elements of normal skin. These findings are important for correct interpretation of staining patterns, when S100 antibodies are used as markers for melanoma or other tumours.

Most applications of thoracic EIT use a single plane of electrodes on the chest from which a transverse image 'slice' is calculated. However, interpretation of EIT images is made difficult by the large region above and below the electrode plane to which EIT is sensitive. Volumetric EIT images using two (or more) electrode planes should help compensate, but are little used currently. The Graz consensus reconstruction algorithm for EIT (GREIT) has become popular in lung EIT. One shortcoming of the original formulation of GREIT is its restriction to reconstruction onto a 2D planar image. We present an extension of the GREIT algorithm to 3D and develop open-source tools to evaluate its performance as a function of the choice of stimulation and measurement pattern. Results show 3D GREIT using two electrode layers has significantly more uniform sensitivity profiles through the chest region. Overall, the advantages of 3D EIT are compelling.

Supplemental Digital Content is available in the text