Patients, clinical data, and pathological tissue sample collection
This study was approved by the review board of each hospital, including Kyoto University (confirmation No. R1230), and patients provided informed consent for sample use and data analysis through opt-out agreements. This multicenter, retrospective observational study was conducted at Kyoto University, Osaka Red Cross Hospital, Kyoto Medical Center, Kyoto Katsura Hospital, Tenri Hospital, and Otsu Municipal Hospital. The study protocol was approved by the ethics committee of Kyoto University and each hospital. This study was conducted in accordance with the Ethical Guidelines for Medical and Health Research Involving Human Subjects and the Declaration of Helsinki.
We retrospectively enrolled 174 patients who received systemic Tmab-CTx for HER2-GEA between January 2011 and December 2016 in any of the participating hospitals. Clinical data and pathological tissue samples were collected from patients at each hospital. We included patients 1) whose HER2 score was 2 plus and the fluorescence in situ hybridization result was positive or whose HER2 score was 3 plus, and 2) who had received Tmab-CTx. The exclusion criteria were as follows: 1) pathological tissue samples or detailed clinical information could not be obtained, and 2) could not be evaluated for the objective response to chemotherapy. Using the database of each hospital, we obtained the clinical information of these patients since they received Tmab-CTx retrospectively. Comorbidity was assessed using the CCI.
Pathological tissue samples with formalin-fixed, paraffin-embedded sections were obtained from the specimens of surgeries or endoscopic biopsy when the case was diagnosed as HER2-GEA. These samples were processed at the pathological department of each hospital.
PTEN expression analysis
Tissue sections were analyzed for PTEN by IHC with PTEN monoclonal antibody (clone 138G6, #9559, Cell Signaling Technology, Danvers, MA; diluted 1:200) at Kyoto University, as reported previously 13,33. PTEN IHC was subjectively scored as absent (0) without detectable immunostaining in the whole cancer specimen, as weak (1+) with low cytoplasmic staining, as moderate (2+) with intermediate staining (between weak and strong), and as strong (3+) with intense staining (Supplementary Fig. 3). PTEN loss was defined as negative staining (score 0) of cells in more than 75 % of the tumor 13. Interpretation was performed by three independent observers, including a specialist of pathology, and discrepancies were discussed to obtain a final result.
Clinical response to Tmab-CTx
We evaluated the clinical objective response to Tmab-CTx according to the RECIST version 1.1 19. The best overall response to Tmab-CTx was recorded as the objective response of each patient. This evaluation was based on radiological images, including computed tomography results, along with magnetic resonance imaging and relevant physical examination findings, all of which were reviewed retrospectively in a blinded manner by at least two clinicians.
The sum of the proportion of complete response (CR) and partial response (PR) represented the response rate used to compare the PTEN-loss with the PTEN-positive group. Then, the DCR, which was defined as the sum of the proportions of CR, PR, SD, and non-CR/non-PD, was used as the study’s primary endpoint for comparing these two groups. In addition, DCR was evaluated in patients with target lesions. The duration of SD, which represented the period from the Tmab-CTx initiation date to the progressive disease (PD) diagnosis date, was also evaluated and compared between the two groups.
OS and PFS
OS was defined as the period from the Tmab-CTx initiation date to the date of the patient’s death. PFS was defined as the period from the Tmab-CTx initiation date to the date of disease “progression” onset based on an objective evaluation of the patient chart or to the date of the patient’s death.
Cell lines, cell culture, and reagents
Human gastric adenocarcinoma cell line NCI-N87 (RRID: CVCL_1603) was purchased from the American Type Culture Collection (Manassas, VA). Human esophageal adenocarcinoma cell line OE19 (RRID: CVCL_1622) was purchased from the European Collection of Cell Cultures (Salisbury, UK). These cell lines were confirmed by STR profiling, and were mycoplasma-free. Cell lines were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum, penicillin (100 U/mL), and streptomycin (100 μg/mL; Life Technologies) and were incubated at 37°C in a humidified chamber containing 5% CO2. Tmab was provided by Chugai Pharmaceutical (Tokyo, Japan) for nonclinical investigations; 5-FU was purchased from Wako (Tokyo, Japan); CDDP was provided by YAKULT HONSHA (Tokyo, Japan). LY294002 and NVP-BEZ235 were purchased from Cayman Chemical (Ann Arbor, MA). Everolimus (RAD001) and MK2206 were purchased from Selleck Chemicals (Houston, TX).
Small interfering RNA and short hairpin RNA
Two distinct siRNA species targeting PTEN (siPTEN #1, Hs_PTEN_6 FlexiTube siRNA, SI00301504; siPTEN #2, Hs_PTEN_8 FlexiTube siRNA, SI03048178) and nonsilencing control siRNA (AllStars negative control siRNA, SI03650318) were purchased from Qiagen and transfected with Lipofectamine RNAiMAX (Invitrogen Life Technologies, Carlsbad, CA) according to the manufacturer’s protocol.
The shRNA vector (PTEN Human shRNA Plasmid Kit, TL320498) was purchased from Origene (Rockville, MA). The Plasmid kit contained a non-effective 29-mer scrambled shRNA and four unique 29-mer shRNA constructs, #A #B #C #D, in the lentiviral GFP vector, from which we selected #C and #D shRNA because they had better PTEN knockdown efficacy. The packaging vector (psPAX2, plasmid 12260) and envelope vector (pMD2.G, plasmid 12259) were purchased from Addgene (Cambridge, MA). The shRNA lentivirus containing shRNA constructs were prepared according to the manufacturer’s protocol, and cell lines were infected with them. We selected stable knockdown clones by flow cytometry to collect GFP-positive cells.
Western blotting
Cells were lysed in sodium dodecyl sulfate lysis buffer supplemented with a protease inhibitor cocktail (Nacalai Tesque, Kyoto, Japan) and phosphatase inhibitor cocktail (Nacalai Tesque, Kyoto, Japan). A total of 20 μg of whole-cell lysate was subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride membrane (Merck Millipore). Membranes were probed with specific primary antibodies against HER2 (polyclonal, #2242), PTEN (clone 138G6, #9559), pan-Akt (clone C67E7, #4691), phosphorylated Akt (Ser473, clone D9E, #4060), p44/p42 mitogen-activated protein kinase [MAPK; ERK1/2] (clone 137F5, #4695), phosphorylated p44/p42 MAPK (ERK1/2) (Thr202/Tyr204, clone D13.14.4E, #4370), S6 ribosomal protein (clone 5G10, #2217), phosphorylated S6 ribosomal protein (Ser235/236, clone D57.2.2E, #4858) (Cell Signaling Technology), and horseradish peroxidase (HRP)-conjugated secondary antibody (Dako, Santa Clara, CA). HRP-conjugated β-actin antibody (Sigma-Aldrich, St. Louis, MO) was used as a loading control. Bands were visualized using a Pierce Western blotting substrate kit (Thermo Scientific).
Cell viability and cell growth inhibition assay
Cell viability was measured by the WST-8 colorimetric assay using the Cell Counting Kit-8 (CCK-8; Dojindo, Kumamoto, Japan). Exponentially growing cells (2500–3500/100 μL/well) were seeded in triplicate into 96-well plates and cultured in reagent-containing medium for 120 h. Then, 10 μL of CCK-8 was added to each well, and incubation continued for 3 h. The absorbance at 450nm was measured with a GloMax-Multi detection system (Promega, Madison, WI) to calculate the number of viable cells per well. The working concentrations of the reagents were in the range in which the tumor growth inhibition rate was 30%–50%. We investigated the tumor inhibition rate of each reagent in the original cell lines at various concentrations (Supplementary Fig. 4). The optimized reagent concentrations in NCl-N87 and OE19 cell cultures were derived from preliminary experiments: Tmab, 10 µg/mL in both cell lines; 5-FU, 1 and 10 µM in NCl-N87 and OE19, respectively; CDDP, 1 µM in both cell lines; LY294002, 5 and 10 µM in NCl-N87 and OE19, respectively; everolimus, 10 nM in both cell lines; MK-2206, 500 nM and 5 µM in NCl-N87 and OE19, respectively; NVP-BEZ235, 50 and 500 nM in NCl-N87 and OE19, respectively.
Statistical analysis
All values are expressed as mean ± standard deviation. All in vitro experiments were repeated at least three times. Categorical data were analyzed using Fisher’s exact test. Continuous variables were analyzed using Student’s t-test. To investigate factors associated with OS or PFS, multivariate logistic regression analysis was used for factors that were included in the model with p<0.20 in univariate analysis. Survival curves were calculated according to the Kaplan-Meier method and analyzed using the log-rank test. All analyses were two-sided, and differences with p<0.05 were considered statistically significant. Statistical analyses were performed using JUMP software, version 14.0, Statistical Discovery (SAS, Cary, NC).