A pipeline for evaluation of machine learning/AI models to quantify PD-L1 immunohistochemistry.

Immunohistochemistry (IHC) is used to guide treatment decisions in multiple cancer types. For treatment with checkpoint inhibitors, PD-L1 IHC is used as a companion diagnostic. However, the scoring of PD-L1 is complicated by its expression in cancer and immune cells. Separation of cancer and non-cancer regions is needed to calculate tumor proportion scores (TPS) of PD-L1, which is based on the percentage of PD-L1 positive cancer cells. Evaluation of PD-L1 expression requires highly experienced pathologists and is often challenging and time consuming. Here we used a multi-institutional cohort of 77 lung cancer cases stained centrally with the PD-L1 22C3 clone. We developed a four-step pipeline for measuring TPS that includes the co-registration of H&E, PD-L1 and negative control (NC) digital slides for exclusion of necrosis, segmentation of cancer regions and quantification of PD-L1+ cells. As cancer segmentation is a challenging step for TPS generation, we trained DeepLab V3 in the Visiopharm software package to outline cancer regions in PD-L1 and negative control (NC) images and evaluated the model performance by mean intersection over union (mIoU) against manual outlines. Only 14 cases were required to accomplish an mIoU of 0.82 for cancer segmentation in hematoxylin stained NC cases. For PD-L1 stained slides, a model trained on PD-L1 tiles augmented by registered NC tiles achieved an mIoU of 0.79. In segmented cancer regions from whole slide images, the digital TPS achieved an accuracy of 75% against the manual TPS scores from the pathology report. Major reasons for algorithmic inaccuracies include the inclusion of immune cells in cancer outlines and poor nuclear segmentation of cancer cells. Our transparent and stepwise approach and performance metrics can be applied to any IHC assay to provide pathologists with important insights when to apply and how to evaluate commercial automated IHC scoring systems.

Third-party fecal microbiota transplantation for high-risk treatment-naïve acute GVHD of the lower GI tract.

Disruption of the intestinal microbiome is observed with acute graft-versus-host disease (GVHD) of the lower gastrointestinal (LGI) tract and fecal microbiota transplantation (FMT) has successfully cured steroid-refractory cases. In this open-label, single-arm, pilot study (NCT04139577), third-party, single donor FMT was administered in combination with systemic corticosteroids to participants with high-risk acute LGI GVHD, with a focus on treatment-naïve cases. Participants were scheduled to receive one induction dose (15 capsules/day for 2 consecutive days), followed by 3 weekly maintenance doses, consisting of 15 capsules/dose. The primary endpoint of the study was feasibility, which would be achieved if ≥80% of participants able to swallow ≥40 of the 75 scheduled capsules. Ten participants (9 treatment-naïve; 1 steroid-refractory) were enrolled and treated. The study met the primary endpoint, with 9 of 10 participants completing all eligible doses. Organ-specific LGI complete response rate at Day 28 was 70%. Initial clinical response was observed within 1 week for all responders and clinical responses were durable, without recurrent LGI GVHD in complete responders. Exploratory analyses suggest that alpha diversity increased following FMT. While recipient microbiome composition never achieved a high degree of donor similarity, expansion of donor-derived species and increases in tryptophan metabolites and short-chain fatty acids were observed within the first 7 days after FMT. Investigation into the use of microbiome-targeted interventions earlier in the treatment paradigm for acute LGI GVHD is warranted.

Mitigating the attachment of Salmonella Infantis on isolated poultry skin with cetylpyridinium chloride.

To provide the poultry industry with effective mitigation strategies, the effects of cetylpyridinium chloride (CPC) on the reduction of Salmonella Infantis, hilA expression, and chicken skin microbiota were evaluated. Chicken breast skins (4×4 cm; N = 100, n = 10, k = 5) were inoculated with Salmonella (Typhimurium or Infantis) at 4°C (30min) to obtain 108 CFU/g attachment. Skins were shaken (30s), with remaining bacteria being considered firmly attached. Treatments were applied as 30s dips in 50 mL: no inocula-no-treatment control (NINTC), no treatment control (NTC), tap water (TW), TW+600 ppm PAA (PAA), or TW+0.5% CPC (CPC). Excess fluid was shaken off (30s). Samples were homogenized in nBPW (1 min). Samples were discarded. Salmonella was enumerated and Log10 transformed. Reverse transcriptase-qPCR (rt-qPCR) was performed targeting hilA gene and normalized using the 2-ΔΔCt method. Data were analyzed using one-way ANOVA in RStudio with means separated by Tukey's HSD (P≤0.05). Genomic DNA of rinsates was extracted, 16S rRNA gene (V4) was sequenced (MiSeq), and data analyzed in QIIME2 (P≤0.05 and Q≤0.05). CPC and PAA affected Salmonella levels differently with CPC being effective against S. Infantis compared to TW (P<0.05). Treatment with CPC on S. Infantis-infected skin altered the hilA expression compared to TW (P<0.05). When inoculated with S. Typhimurium, there was no difference between the microbiota diversity of skins treated with PAA and CPC; however, when inoculated with S. Infantis, there was a difference in the Shannon's Entropy and Jaccard Dissimilarity between the two treatments (P<0.05). Using ANCOM at the genus level, Brochothrix was significant (W = 118) among skin inoculated with S. Typhimurium. Among S. Infantis inoculated, Yersiniaceae, Enterobacterales, Lachnospiraceae CHKCI001, Clostridia vadinBB60 group, Leuconostoc, Campylobacter, and bacteria were significant (408). CPC and PAA-treated skins had lowest relative abundance of the genera. In conclusion, CPC mitigated Salmonella Infantis, altered hilA expression, and influenced the chicken skin microbiota.

Multicenter Phase I Trial of Ivosidenib as Maintenance Treatment Following Allogeneic Hematopoietic Cell Transplantation for IDH1-Mutated Acute Myeloid Leukemia.

PURPOSE: Isocitrate dehydrogenase 1 (IDH1) mutations occur in 5% to 10% of patients with acute myeloid leukemia (AML). Ivosidenib is an IDH1 inhibitor, approved for use in patients with IDH1-mutated AML. PATIENTS AND METHODS: We conducted a multicenter, phase I trial of maintenance ivosidenib following allogeneic hematopoietic cell transplantation (HCT) in patients with IDH1-mutated AML. Ivosidenib was initiated between days 30 and 90 following HCT and continued for up to 12 28-day cycles. The first dose level was 500 mg daily, with level reduction to 250 mg daily, if needed, in a 3 × 3 de-escalation design. Ten additional patients would then receive the MTD or recommended phase 2 dose (RP2D). The primary endpoint was establishing the MTD or RP2D of ivosidenib. RESULTS: Eighteen patients were enrolled, of whom 16 initiated post-HCT ivosidenib. One dose-limiting toxicity, grade(g) 3 QTc prolongation, was observed. The RP2D was established at 500 mg daily. Attributable g≥3 adverse events were uncommon, with the most common being QTc prolongation in 2 patients. Eight patients discontinued maintenance, with only one due to adverse event. Six-month cumulative incidence (CI) of gII-IV aGVHD was 6.3%, and 2-year CI of all cGVHD was 63%. Two-year CI of relapse and nonrelapse mortality (NRM) were 19% and 0%, respectively. Two-year progression-free (PFS) was 81%, and 2-year overall survival (OS) was 88%. CONCLUSIONS: Ivosidenib is safe and well-tolerated as maintenance therapy following HCT. Cumulative incidence of relapse and NRM, as well as estimations of PFS and OS, were promising in this phase I study.

Enasidenib as maintenance following allogeneic hematopoietic cell transplantation for IDH2-mutated myeloid malignancies.

IDH2 (isocitrate dehydrogenase 2) mutations occur in approximately 15% of patients with acute myeloid leukemia (AML). The IDH2 inhibitor enasidenib was recently approved for IDH2-mutated relapsed or refractory AML. We conducted a multi-center, phase I trial of maintenance enasidenib following allogeneic hematopoietic cell transplantation (HCT) in patients with IDH2-mutated myeloid malignancies. Two dose levels, 50mg and 100mg daily were studied in a 3 × 3 dose-escalation design, with 10 additional patients treated at the recommended phase 2 dose (RP2D). Enasidenib was initiated between days 30 and 90 following HCT and continued for twelve 28-day cycles. Twenty-three patients were enrolled, of whom 19 initiated post-HCT maintenance. Two had myelodysplastic syndrome, and 17 had AML. All but 3 were in first complete remission. No dose limiting toxicities were observed, and the RP2D was established at 100mg daily. Attributable grade ≥3 toxicities were rare, with the most common being cytopenias. Eight patients stopped maintenance before completing 12 cycles, due to adverse events (n=3), pursuing treatment for graft-vs-host disease (GVHD) (n=2), clinician choice (n=1), relapse (n=1), and COVID infection (n=1). No cases of grade ≥3 acute GVHD were seen, and 12-month cumulative incidence of moderate/severe chronic GVHD was 42% (20-63%). Cumulative incidence of relapse was 16% (95% CI: 3.7-36%); 1 subject relapsed while receiving maintenance. Two-year progression-free and overall survival were 69% (95% CI: 39-86%) and 74% (95% CI, 44-90%), respectively. Enasidenib is safe, well-tolerated, with preliminary activity as maintenance therapy following HCT, and merits additional study. The study was registered at www.clinicaltrials.gov (#NCT03515512).

Whipple's disease: a fatal mimic.

Whipple's Disease, a rare diagnosis caused by the slow-growing bacterium Tropheryma whipplei, most often presents with the classically described signs of malabsorption due to gastrointestinal colonization. However, it can also have signs and symptoms that clinically overlap with rheumatic diseases, potentially resulting in misdiagnosis. Furthermore, treatment with modern potent biologic immunosuppressive agents and classic disease modifying anti-rheumatic drugs (DMARDs) can lead to serious exacerbation of undiagnosed infections. We present the case of a middle-aged woman with long term complaints of arthalgias, who was diagnosed with seronegative rheumatoid arthritis and subsequently treated for almost 7 years with such immunosuppressive therapies. The patient's disease course included chronic diarrhea that abruptly intensified and culminated in fatal hypovolemic shock/sepsis. A diagnosis of WD was made by autopsy examination, wherein several organ systems were found to be heavily involved by Tropheryma whipplei organisms, and their identification was confirmed with histochemical and molecular evaluation. Notably, most bacterial organisms were located deeply in the submucosa/muscularis of affected organs, a practical reminder to practicing pathologists that challenges the classic histopathologic description of Whipple disease as an infiltration of predominantly lamina propria, and the potential for sampling bias in typically superficial endoscopic biopsies during routine procedures.

Whipple's disease: a fatal mimic

Whipple's Disease, a rare diagnosis caused by the slow-growing bacterium Tropheryma whipplei, most often presents with the classically described signs of malabsorption due to gastrointestinal colonization. However, it can also have signs and symptoms that clinically overlap with rheumatic diseases, potentially resulting in misdiagnosis. Furthermore, treatment with modern potent biologic immunosuppressive agents and classic disease modifying anti-rheumatic drugs (DMARDs) can lead to serious exacerbation of undiagnosed infections. We present the case of a middle-aged woman with long term complaints of arthalgias, who was diagnosed with seronegative rheumatoid arthritis and subsequently treated for almost 7 years with such immunosuppressive therapies. The patient's disease course included chronic diarrhea that abruptly intensified and culminated in fatal hypovolemic shock/sepsis. A diagnosis of WD was made by autopsy examination, wherein several organ systems were found to be heavily involved by Tropheryma whipplei organisms, and their identification was confirmed with histochemical and molecular evaluation. Notably, most bacterial organisms were located deeply in the submucosa/muscularis of affected organs, a practical reminder to practicing pathologists that challenges the classic histopathologic description of Whipple disease as an infiltration of predominantly lamina propria, and the potential for sampling bias in typically superficial endoscopic biopsies during routine procedures.