The PAT-resistant patient was selected from HER2-positive breast cancer patients (all >40 years) who were treated by mastectomy at the Mutsu General Hospital (Aomori, Japan) between February, 2017 and October, 2019. During this research period, a PAT-resistant case was found and further analysis was performed. The characteristic treatment flow and investigational points are summarized in Fig. 1A. The diagnosis was ductal adenocarcinoma in primary and solid tubular et scirrhous carcinoma in recurrent, T_4bN_2aM₀, Stage IIIB. The patient was treated using Fluorouracil + Epirubicin Hydrochloride + Cyclophosphamide Hydrate (PAT1; day 18-78), and then with Trastuzumab + Paclitaxel (PAT2; day 102-466) (Fig. 1A). External radiotherapy (photon beam, 50 Gy per fraction, 25 fractions) was carried out on day 494-529 (PAT3). After surgery for removal of recurrent cancer tissue, the patient was treated using Trastuzumab + Pertuzumab + Docetaxel (PAT4; from day 669).

The study was approved by the Committee of Medical Ethics of the Hirosaki University Graduate School of Health Sciences (Hirosaki, Japan; no. 2016-051) and Mutsu General Hospital (Mutsu, Japan; no. 000-282) to ensure the welfare and privacy of the donors. Following a detailed verbal explanation regarding the content of this study, a written informed consent was obtained.

To observe the patient's condition during the treatment, counts of white blood cells (WBC), red blood cells (RBC) and platelets (Plt) were recorded and the level of hemoglobin (Hg) was measured by routine hematological techniques. At the same time, the concentration of C-reactive protein (CRP), carcinoembryonic antigen (CEA), cancer antigen 15-3 (CA15-3) and breast cancer antigen 225 (BCA225) were also quantified.

PB was collected directly into serum separation tubes (BD Biosciences). The serum was separated and stored at -80˚C.

Serum metabolites were measured as part of a targeted metabolomics panel using the AbsoluteIDQ P180 kit (Biocrates, Inc.). The kit is a fully automated assay based on phenylisothiocyanate (PITC) derivatization of the target analytes using internal standards for quantitation. Human serum sample preparation was carried out according to the manufacturer's protocol. Briefly, 10 µl of serum was transferred to the first 96-well plate and dried under a nitrogen stream. Thereafter, 50 µl of a 5% PITC solution was added to derivatize amino acids and biogenic amines. After incubation, the filter spots were dried again before the metabolites were extracted using 5 mM ammonium acetate in methanol (300 µl) into the second 96-well plate for analysis after further dilution using the Biocrates MS running solvent A.

Acylcarnitines, lyso-phosphatidylcholines with acyl residue, phosphatidylcholine with diacyl residue sum (PC aa), and phosphatidylcholine with acyl-alkyl residue sum (PC ae), sphingomyelins, and the sum of hexoses were analyzed by the flow injection analysis (FIA) method in positive ion mode using the liquid chromatography-mass spectrometry (LC-MS/MS). The LC-MS/MS analysis was carried out by an HPLC system (ExionLC™ AD, AB Sciex) coupled to a QTRAP6500+ triple quadruple ion trap hybrid mass spectrometer system (AB Sciex) in electrospray ionization (ESI) mode. The full list of metabolites and their abbreviations are presented in Table I. The metabolites were identified and quantified with multiple reaction monitoring (MRM) with individual transition and parameters of declustering potential (DP) and collision energy (CE) as listed in Table S1. Flow rate settings for FIA were as follows: 0.03 ml/min in 1.6 min; 0.03 to 0.2 ml/min in 0.8 min; 0.2 ml/min in 0.4 min; 0.2 to 0.03 ml/min in 0.2 min. MS settings for FIA mode were as follows: Curtain gas, 45; ion spray voltage, 5,500 V; temperature, 175˚C; ion source gas 1, 40 psi; ion source gas 2, 50 psi; CAD gas, 6 psi; entrance potential, 10 V; collision cell exit potential, 15 V. Twenty microliters of the sample extract were used in the FIA. Quantification was carried out using internal standards and a calibration curve.

Amino acids and biogenic amines were analyzed via LC-MS/MS in positive ion mode. Then, 10 µl of the sample extract were injected onto an HPLC C18 column (Zorbax Eclipse XDB-C18 column, 3x100 mm, 3.5 µm; Agilent) with a guard column (Zorbax SB-C18, 3x100 mm, 1.85 µm; Agilent) at 40˚C using a 10-min solvent gradient employing 0.2% formic acid in water (solvent A) and 0.2% formic acid in acetonitrile (solvent B). Additional LC settings for LC-MS/MS were as follows: 0% B in 0.5 min; 0-95% B in 5 min; 95% B in 1 min; 95-0% B in 0.5 min; 0% B in 2.5 min at a flow rate of 0.5 ml/min. MS settings for LC-MS/MS mode were as follows: Curtain gas, 45; ion spray voltage, 5,500 V; temperature, 500˚C; ion source gas 1, 40 psi; ion source gas 2, 50 psi; CAD gas, 6 psi; entrance potential, 10 V; collision cell exit potential, 15 V. All metabolites were identified and quantified using isotopically-labeled internal standards and MRM as optimized and provided by Biocrates Inc.

Paraffin-embedded sections of the removed cancer tissues were deparaffinized with xylene and ethanol, washed in D-PBS (-) and treated with 3% hydrogen peroxide for 5 min. The slides were treated with 10 mM citrate buffer (pH 6.0) and incubated for 10 min at 121˚C using an autoclave. Then, the slides were washed in TBS buffer (25 mM Tris-HCl and 150 mM NaCl, pH 7.2). To perform blocking, the slides were treated with the blocking solution (5% normal goat serum in TBS buffer) at room temperature for 30 min. Next, the slides were incubated with the blocking solution containing a primary rabbit monoclonal antibody directed against human HER2/ERBB2 (no. 4290; Cell Signaling Technology Inc.) at a 1:1,000 dilution at room temperature for 60 min. The slides were washed three times with TBS buffer and incubated at room temperature for 60 min with an anti-rabbit IgG horseradish peroxidase (HRP)-linked antibody (no. 7074, Cell Signaling Technology) prepared in the blocking solution at a 1:2,000 dilution. The slides were washed three times with TBS buffer at room temperature for 5 min. Signals were visualized using 3,3'-diaminobenzidine (DAB) (Sigma-Aldrich). The slides were washed in water and nuclei were stained using hematoxylin solution. Finally, the slides were dehydrated with ethanol and xylene, and mounted using a hydrophobic mounting agent. Fluorescent immunological analysis was performed as in our previous study (15). Mouse monoclonal antibody directed against human nSMase2 (sc-166637; Santa Cruz Biotechnology, Inc.) (1:100 dilution) and rabbit monoclonal HER2/ERBB2 antibody (1:1,000 dilution) described above was used as a primary antibody. Anti-rabbit IgG Alexa Fluor 488-conjugated antibody (no. 4412, Cell Signaling Technology) and Anti-mouse IgG Alexa Fluor 647-conjugated antibody (no. 4410, Cell Signaling Technology) were used as secondary antibodies. The cell nuclear staining for fluorescence analysis DAPI (no. 4083; Cell Signaling Technology) was used. Fluorescence signals were detected using a fluorescence microscope (BZ-X700; Keyence).

Following diagnosis of right breast adenocarcinoma with the expression of HER2 as primary cancer, the tumor was removed by surgery (day 0, Fig. 1A). Three months after completing PAT3 (day 606), locally recurrent cancer lesion in right breast within radiation field was detected and removed by surgery on day 656. A second locally recurrent cancer lesion and metastatic lymph node tumor were observed on day 911. According to pathological analysis, the removed primary tumor tissue (day 0) showed a distinctly higher expression of HER2 as compared to corresponding areas of normal tissue (Fig. 1B). A transient increase in WBC (days 1 and 657) and Plt (days 7-102 and 669) after surgery was observed (Fig. 2A and B). On the other hand, the number and concentration of RBC was stable (Fig. 2C and D). The inflammation marker CRP responded similarly to WBC and Plt (Fig. 2E), whereas the level of CEA remained stable (Fig. 2F). The levels of CA15-3 and BCA225 gradually increased during the observation period however these markers were within the normal range (Fig. 2G and H).

After the first surgery, 4 cycles of PAT were performed (Fig. 1A). According to pathological analysis, the staging T₁N₁M0 (Stage IIA) at the first diagnostic stage (before day 0) changed to rT_4bN_2aM₀ when the recurrent tumor was detected in the same original region and with a part invading into the greater pectoral muscle as well as presenting as lymph node metastasis (~day 650).

None of the four PAT cycles showed effectiveness. To investigate whether specific biomarker(s) were released from cancer tissues to circulating body fluids, PB serum was collected at the time before surgery for 1st recurrent tissue (day 650, Fig. 1-i), after 13 days from surgery (day 669, Fig. 1-ii) and the day when the 2nd recurrent cancer was diagnosed (day 809, Fig. 1-iii), and analyzed by mass spectrometry for wide-targeted metabolomic markers of response. The concentration of 188 molecules was analyzed (Table I), and, of these, 40 upregulated molecules and 28 downregulated molecules were observed on day 669 (Fig. 1A-ii) as compared to day 650 (Fig. 1A-i) (Fig. 3A and B) by merely calculating the molecules in Table I. In addition, 23 molecules among the 40 upregulated molecules (Fig. 1A-ii) were downregulated on day 809 (Fig. 1A-iii) in comparison to day 669 (Fig. 1A-ii). On the other hand, 23 molecules of the 28 downregulated molecules on day 669 (as compared to day 650) were upregulated on day 809 (as compared to day 669). These molecules that were up-to-down- and down-to-up-regulated contained many lipids of three classes: Acyl-alkyl phosphatidylcholine (6 molecules), sphingomyeline (4 molecules) and lysophosphatidylcholine (lysoPC) (7 molecules) (Fig. 3C). Interestingly, the concentrations of molecules in these three classes showed fluctuations which correlated with the detection of cancer recurrence (Fig. 3C). LysoPC is produced from partial hydrolysis of phosphatidylcholines, and is regarded as an inflammation-resolution lipid mediator (16). In this case study, half of the 14 lysoPC species measured were increased after surgery and decreased when cancer recurred.

Phosphatidylcholines and sphingomyelins are known as components of cell membranes, especially in extracellular vesicles which are associated with cell-to-cell communication (17-19). Therefore, we focused on neutral sphingomyelinase-2 (nSMase), which is one of the important factors for exocytosis or endocytosis. In primary cancer tissue and 1st recurrent cancer tissue, a higher expression of Her2 protein was observed (Fig. 4). However, the lowest accumulation of intracellular nSMase in recurrent tumor was shown in comparison to normal or primary tumor. This means that the exocytosis or endocytosis occurs for release as extracellular vesicles (EVs).