Opal reagents

Human TMAs containing samples from patients were deparaffinized and fixed with 10% neutral buffered formalin. Antigen retrieval was performed using citrate buffer (pH 6.0). Slides were then blocked with 10% BSA + 0.05% Tween20 and the antibodies were diluted in 2% BSA in 0.05% PBST and used in a multiplexed manner with OPAL reagents (Akoya Biosciences). All primary antibodies (described in Supplementary Table S4) were incubated overnight at 4 °C (1:400) according to the following staining sequences: Set 1: KLC1 → KIF5B → CK → DAPI; Set 2: EIF4G2 → CK → DAPI. Each antibody was validated and optimized separately and then the multiplex staining was optimized. Briefly, following primary antibody incubation, slides were washed with 0.05% PBST, incubated with secondary antibodies conjugated to HRP (1:400), washed again, and incubated with OPAL reagents. Slides were then washed, and antigen retrieval was performed. Then, slides were washed with PBS and stained with the next primary antibody or with DAPI at the end of the cycle. Finally, slides were mounted using Immu-mount (#9990402, Thermo Scientific). Images were taken with a Phenocycler scanner (Akoya®) and analyzed using QuPath software. Cell segmentation was done using Cellpose. For each patient, average cell intensities for the indicated markers was calculated, and correlation between the different markers was analyzed.

Multiplex immunofluorescence was performed using OPAL reagents (Akoya Biosciences, Marlborough, MA, USA) on 4-μm sections of FFPE tumor biopsies as described previously (18 ). The sequence of antibody stainings was as follows: 1. CD4 (FP1600/EP204, Akoya Biosciences, 1:100) – OPAL520; 2. CD8 (FP1601/144B, Akoya Biosciences, 1:200) – OPAL690; 3. CD66b (80H3, Sanbio, 1:100) – OPAL570; 4. Cytokeratin-Pan (AE1/AE3, Invitrogen, 1:500) – Opal 620; and 5. DAPI. Digital image analysis was also performed as described previously (18 ). Cellular densities were calculated by dividing the number of cells with a certain phenotype by the total area of that region and were averaged per patient across all regions of interest. Distances from a cell with a certain phenotype to the nearest cell with another phenotype were calculated by nearest neighbor analysis; this was averaged across all cells, and per patient across all regions of interest.

Multiplexed IF was performed on 4-μm trephine bone marrow sections from an exploratory patient cohort from the HOVON87/NMSG18 trial using antibodies against distinct immune cell subsets/markers in combination with OPAL reagents (Akoya Biosciences, Marlboro, MA, USA) (for specifics of antibodies and reagents, see Table S1). Patients (n = 22) were divided into two groups with high- and low OS using the median value as a cut-off. Immune staining included multiple cycles of: antigen retrieval (15 min boiling in antigen retrieval buffer, pH 6 or pH 9 depending on primary antibodies) followed by cooling, blocking, and consecutive staining with primary antibodies, HRP-polymer, and Opal fluorophores; cycles were repeated until all markers were stained. Finally, nuclei were stained with DAPI.

Multiplexed IF was performed using OPAL reagents (Akoya Biosciences) on whole slides (using a randomly selected subset of cohort A with comparable fractions of all spatial phenotypes). In brief, stainings included multiple cycles of antigen retrieval (15 min boiling in antigen retrieval buffer, pH 6 or pH 9 depending on primary antibodies) followed by cooling, blocking, and consecutive staining with primary antibodies, HRP-polymer, and Opal fluorophores; cycles were repeated until all markers were stained. Finally, nuclei were stained with DAPI.