Kinase Inhibitor Treatment of Patients with Advanced Cancer Results in High Tumor Drug Concentrations and in Specific Alterations of the Tumor Phosphoproteome

Longitudinal phosphoproteomics reveals the PI3K-PAK1 axis as a potential target for recurrent colorectal liver metastases

The resistance of colorectal cancer liver metastases (CRLMs) to 5-fluorouracil (5-FU) chemotherapy remains a significant global health challenge. We investigated the phosphoproteomic dynamics of serial tissue sections obtained from initial metastases and recurrent tumors collected from 24 patients to address this unmet need for innovative therapeutic strategies for patients with CRLM with a poor prognosis. Our analysis revealed the activation of PAK kinase in patients with CRLM with a poor prognosis. Using an unbiased computational approach, we conducted a correlation analysis between PAK1 kinase activity and 545 drug sensitivity profiles across 35 colorectal cancer cell lines and identified PI3K inhibitors as potential therapeutic candidates. The efficacy of the FDA-approved PI3K inhibitor copanlisib was validated in 5-FU-resistant cell lines with high PAK1 kinase activity both in vitro and in vivo. This study presents an effective strategy for drug target discovery based on kinase activity, and the concept of this approach is widely applicable.

The STELLAR trial: a phase II/III randomized trial of high-dose, intermittent sunitinib in patients with recurrent glioblastoma

Previously, the tyrosine kinase inhibitor sunitinib failed to show clinical benefit in patients with recurrent glioblastoma. Low intratumoural sunitinib accumulation in glioblastoma patients was reported as a possible explanation for the lack of therapeutic benefit. We designed a randomized phase II/III trial to evaluate whether a high-dose intermittent sunitinib schedule, aimed to increase intratumoural drug concentrations, would result in improved clinical benefit compared to standard treatment with lomustine. Patients with recurrent glioblastoma were randomized 1:1 to high-dose intermittent sunitinib 300 mg once weekly (Q1W, part 1) or 700 mg once every two weeks (Q2W, part 2) or lomustine. The primary end-point was progression-free survival. Based on the pre-planned interim analysis, the trial was terminated for futility after including 26 and 29 patients in parts 1 and 2. Median progression-free survival of sunitinib 300 mg Q1W was 1.5 months (95% CI 1.4-1.7) compared to 1.5 months (95% CI 1.4-1.6) in the lomustine arm (P = 0.59). Median progression-free survival of sunitinib 700 mg Q2W was 1.4 months (95% CI 1.2-1.6) versus 1.6 months (95% CI 1.3-1.8) for lomustine (P = 0.70). Adverse events (≥grade 3) were observed in 25%, 21% and 31% of patients treated with sunitinib 300 mg Q1W, sunitinib 700 mg Q2W and lomustine, respectively (P = 0.92). To conclude, high-dose intermittent sunitinib treatment failed to improve the outcome of patients with recurrent glioblastoma when compared to standard lomustine therapy. Since lomustine remains a poor standard treatment strategy for glioblastoma, innovative treatment strategies are urgently needed.

Inferring kinase activity from phosphoproteomic data: Tool comparison and recent applications

Aberrant cellular signaling pathways are a hallmark of cancer and other diseases. One of the most important signaling mechanisms involves protein phosphorylation/dephosphorylation. Protein phosphorylation is catalyzed by protein kinases, and over 530 protein kinases have been identified in the human genome. Aberrant kinase activity is one of the drivers of tumorigenesis and cancer progression and results in altered phosphorylation abundance of downstream substrates. Upstream kinase activity can be inferred from the global collection of phosphorylated substrates. Mass spectrometry-based phosphoproteomic experiments nowadays routinely allow identification and quantitation of >10k phosphosites per biological sample. This substrate phosphorylation footprint can be used to infer upstream kinase activities using tools like Kinase Substrate Enrichment Analysis (KSEA), Posttranslational Modification Substrate Enrichment Analysis (PTM-SEA), and Integrative Inferred Kinase Activity Analysis (INKA). Since the topic of kinase activity inference is very active with many new approaches reported in the past 3 years, we would like to give an overview of the field. In this review, an inventory of kinase activity inference tools, their underlying algorithms, statistical frameworks, kinase-substrate databases, and user-friendliness is presented. The most widely-used tools are compared in-depth. Subsequently, recent applications of the tools are described focusing on clinical tissues and hematological samples. Two main application areas for kinase activity inference tools can be discerned. (1) Maximal biological insights can be obtained from large data sets with group comparisons using multiple complementary tools (e.g., PTM-SEA and KSEA or INKA). (2) In the oncology context where personalized treatment requires analysis of single samples, INKA for example, has emerged as tool that can prioritize actionable kinases for targeted inhibition.

High-dose short-term osimertinib treatment is effective in patient-derived metastatic colorectal cancer organoids

BACKGROUND

Most tyrosine kinase inhibitors (TKIs) have failed in clinical trials for metastatic colorectal cancer (mCRC). To leverage the additional lower-affinity targets that most TKIs have, high-dose regimens that trigger efficacy are explored. Here, we studied unprecedented drug exposure-response relationships in vitro using mCRC patient-derived tumour organoids (PDTOs).

METHODS

We investigated the cytotoxic anti-tumour effect of high-dose, short-term (HDST) TKI treatment on 5 PDTOs. Sunitinib, cediranib and osimertinib were selected based on favourable physicochemical and pharmacokinetic properties. Intra-tumoroid TKI concentrations were measured using a clinically validated LC/MS-MS method. Cell death was determined using an enzyme activity assay, immunofluorescent staining and western blotting.

RESULTS

Most PDTOs tested were sensitive to sunitinib and cediranib, but all to osimertinib. Furthermore, HDST osimertinib treatment effectively blocks organoid growth. This treatment led to markedly elevated intra-tumoroid TKI concentrations, which correlated with PDTO sensitivity. Mechanistically, HDST osimertinib treatment induced apoptosis in treated PDTOs.

CONCLUSION

Our work provides a better understanding of TKI exposure vs response and can be used to determine patient-specific sensitivity. Additionally, these results may guide both mechanistic elucidation in organotypic translational models and the translation of target drug exposure to clinical dosing strategies. Moreover, HDST osimertinib treatment warrants clinical exploration for mCRC.

Candidate biomarkers for treatment benefit from sunitinib in patients with advanced renal cell carcinoma using mass spectrometry-based (phospho)proteomics

The tyrosine kinase inhibitor sunitinib is an effective first-line treatment for patients with advanced renal cell carcinoma (RCC). Hypothesizing that a functional read-out by mass spectrometry-based (phospho, p-)proteomics will identify predictive biomarkers for treatment outcome of sunitinib, tumor tissues of 26 RCC patients were analyzed. Eight patients had primary resistant (RES) and 18 sensitive (SENS) RCC. A 78 phosphosite signature (p < 0.05, fold-change > 2) was identified; 22 p-sites were upregulated in RES (unique in RES: BCAR3, NOP58, EIF4A2, GDI1) and 56 in SENS (35 unique). EIF4A1/EIF4A2 were differentially expressed in RES at the (p-)proteome and, in an independent cohort, transcriptome level. Inferred kinase activity of MAPK3 (p = 0.026) and EGFR (p = 0.045) as determined by INKA was higher in SENS. Posttranslational modifications signature enrichment analysis showed that different p-site-centric signatures were enriched (p < 0.05), of which FGF1 and prolactin pathways in RES and, in SENS, vanadate and thrombin treatment pathways, were most significant. In conclusion, the RCC (phospho)proteome revealed differential p-sites and kinase activities associated with sunitinib resistance and sensitivity. Independent validation is warranted to develop an assay for upfront identification of patients who are intrinsically resistant to sunitinib.

Phosphoproteomics of patient-derived xenografts identifies targets and markers associated with sensitivity and resistance to EGFR blockade in colorectal cancer

Epidermal growth factor receptor (EGFR) is a well-exploited therapeutic target in metastatic colorectal cancer (mCRC). Unfortunately, not all patients benefit from current EGFR inhibitors. Mass spectrometry-based proteomics and phosphoproteomics were performed on 30 genomically and pharmacologically characterized mCRC patient-derived xenografts (PDXs) to investigate the molecular basis of response to EGFR blockade and identify alternative drug targets to overcome resistance. Both the tyrosine and global phosphoproteome as well as the proteome harbored distinctive response signatures. We found that increased pathway activity related to mitogen-activated protein kinase (MAPK) inhibition and abundant tyrosine phosphorylation of cell junction proteins, such as CXADR and CLDN1/3, in sensitive tumors, whereas epithelial-mesenchymal transition and increased MAPK and AKT signaling were more prevalent in resistant tumors. Furthermore, the ranking of kinase activities in single samples confirmed the driver activity of ERBB2, EGFR, and MET in cetuximab-resistant tumors. This analysis also revealed high kinase activity of several members of the Src and ephrin kinase family in 2 CRC PDX models with genomically unexplained resistance. Inhibition of these hyperactive kinases, alone or in combination with cetuximab, resulted in growth inhibition of ex vivo PDX-derived organoids and in vivo PDXs. Together, these findings highlight the potential value of phosphoproteomics to improve our understanding of anti-EGFR treatment and response prediction in mCRC and bring to the forefront alternative drug targets in cetuximab-resistant tumors.

Measuring Tumour Imatinib Concentrations in Gastrointestinal Stromal Tumours: Relevant or Redundant?

Imatinib plasma trough concentrations are associated with efficacy for patients treated for advanced or metastatic KIT-positive gastrointestinal stromal tumours (GISTs). This relationship has not been studied for patients treated in the neoadjuvant setting, let alone its correlation with tumour drug concentrations. In this exploratory study we aimed to determine the correlation between plasma and tumour imatinib concentrations in the neoadjuvant setting, investigate tumour imatinib distribution patterns within GISTs, and analyse its correlation with pathological response. Imatinib concentrations were measured in both plasma and in three regions of the resected primary tumour: the core, middle part, and periphery. Twenty-four tumour samples derived from the primary tumours of eight patients were included in the analyses. Imatinib tumour concentrations were higher compared to plasma concentrations. No correlation was observed between plasma and tumour concentrations. Interpatient variability in tumour concentrations was high compared to interindividual variability in plasma concentrations. Although imatinib accumulates in tumour tissue, no distribution pattern of imatinib in tumour tissue could be identified. There was no correlation between imatinib concentrations in tumour tissue and pathological treatment response.

Advancing wide implementation of precision oncology: A liquid nitrogen-free snap freezer preserves molecular profiles of biological samples

PURPOSE

In precision oncology, tumor molecular profiles guide selection of therapy. Standardized snap freezing of tissue biospecimens is necessary to ensure reproducible, high-quality samples that preserve tumor biology for adequate molecular profiling. Quenching in liquid nitrogen (LN₂ ) is the golden standard method, but LN₂ has several limitations. We developed a LN₂ -independent snap freezer with adjustable cold sink temperature. To benchmark this device against the golden standard, we compared molecular profiles of biospecimens.

METHODS

Cancer cell lines and core needle normal tissue biopsies from five patients' liver resection specimens were used to compare mass spectrometry (MS)-based global phosphoproteomic and RNA sequencing profiles and RNA integrity obtained by both freezing methods.

RESULTS

Unsupervised cluster analysis of phosphoproteomic and transcriptomic profiles of snap freezer versus LN₂ -frozen K562 samples and liver biopsies showed no separation based on freezing method (with Pearson's r 0.96 (range 0.92-0.98) and >0.99 for K562 profiles, respectively), while samples with +2 h bench-time formed a separate cluster. RNA integrity was also similar for both snap freezing methods. Molecular profiles of liver biopsies were clearly identified per individual patient regardless of the applied freezing method. Two to 25 s freezing time variations did not induce profiling differences in HCT116 samples.

CONCLUSION

The novel snap freezer preserves high-quality biospecimen and allows identification of individual patients' molecular profiles, while overcoming important limitations of the use of LN₂ . This snap freezer may provide a useful tool in clinical cancer research and practice, enabling a wider implementation of (multi-)omics analyses for precision oncology.

High-Dose Intermittent Treatment with the Multikinase Inhibitor Sunitinib Leads to High Intra-Tumor Drug Exposure in Patients with Advanced Solid Tumors

Patients with advanced cancer refractory to standard treatment were treated with sunitinib at a dose of 300 mg once every week (Q1W) or 700 mg once every two weeks (Q2W). Tumor, skin and plasma concentrations were measured and immunohistochemical staining for tumor cell proliferation (TCP), microvessel density (MVD) and T-cell infiltration was performed on tumor biopsies before and after 17 days of treatment. Oral administration of 300 mg sunitinib Q1W or 700 mg Q2W resulted in 19-fold (range 5-35×) and 37-fold higher (range 10-88×) tumor drug concentrations compared to parallel maximum plasma drug concentrations, respectively. Patients with higher tumor sunitinib concentrations had favorable progression-free and overall survival than those with lower concentrations (p = 0.046 and 0.024, respectively). In addition, immunohistochemistry of tumor biopsies revealed an induction of T-cell infiltration upon treatment. These findings provide pharmacological and biological insights in the clinical benefit from high-dose intermittent sunitinib treatment. It emphasizes the potential benefit from reaching higher tumor drug concentrations and the value of measuring TKI tumor- over plasma-concentrations. The finding that reaching higher tumor drug concentrations provides most clinical benefit in patients with treatment refractory malignancies indicates that the inhibitory potency of sunitinib may be enforced by a high-dose intermittent treatment schedule. These results provide proof of concept for testing other clinically available multitargeted tyrosine kinase inhibitors in a high-dose intermittent treatment schedule.

Tumor Drug Concentration and Phosphoproteomic Profiles After Two Weeks of Treatment With Sunitinib in Patients with Newly Diagnosed Glioblastoma

PURPOSE

Tyrosine kinase inhibitors (TKI) have poor efficacy in patients with glioblastoma (GBM). Here, we studied whether this is predominantly due to restricted blood-brain barrier penetration or more to biological characteristics of GBM.

PATIENTS AND METHODS

Tumor drug concentrations of the TKI sunitinib after 2 weeks of preoperative treatment was determined in 5 patients with GBM and compared with its in vitro inhibitory concentration (IC50) in GBM cell lines. In addition, phosphotyrosine (pTyr)-directed mass spectrometry (MS)-based proteomics was performed to evaluate sunitinib-treated versus control GBM tumors.

RESULTS

The median tumor sunitinib concentration of 1.9 μmol/L (range 1.0-3.4) was 10-fold higher than in concurrent plasma, but three times lower than sunitinib IC50s in GBM cell lines (median 5.4 μmol/L, 3.0-8.5; P = 0.01). pTyr-phosphoproteomic profiles of tumor samples from 4 sunitinib-treated versus 7 control patients revealed 108 significantly up- and 23 downregulated (P < 0.05) phosphopeptides for sunitinib treatment, resulting in an EGFR-centered signaling network. Outlier analysis of kinase activities as a potential strategy to identify drug targets in individual tumors identified nine kinases, including MAPK10 and INSR/IGF1R.

CONCLUSIONS

Achieved tumor sunitinib concentrations in patients with GBM are higher than in plasma, but lower than reported for other tumor types and insufficient to significantly inhibit tumor cell growth in vitro. Therefore, alternative TKI dosing to increase intratumoral sunitinib concentrations might improve clinical benefit for patients with GBM. In parallel, a complex profile of kinase activity in GBM was found, supporting the potential of (phospho)proteomic analysis for the identification of targets for (combination) treatment.

Intra-Tumoral Pharmacokinetics of Pazopanib in Combination with Radiotherapy in Patients with Non-Metastatic Soft-Tissue Sarcoma

There is a lack of understanding whether plasma levels of anticancer drugs (such as pazopanib) correlate with intra-tumoral levels and whether the plasma compartment is the best surrogate for pharmacokinetic and pharmacodynamic evaluation. Therefore, we aimed to quantify pazopanib concentrations in tumor tissue, to assess the correlation between tumor concentrations and plasma concentrations and between tumor concentrations and efficacy. In this clinical trial, non-metastatic STS patients were treated with neo-adjuvant concurrent radiotherapy and pazopanib. Plasma samples and tumor biopsies were collected, and pazopanib concentrations were measured using liquid chromatography-tandem mass spectrometry. Twenty-four evaluable patients were included. The median pazopanib tumor concentration was 19.2 µg/g (range 0.149-200 µg/g). A modest correlation was found between tumor concentrations and plasma levels of pazopanib (ρ = 0.41, p = 0.049). No correlation was found between tumor concentrations and percentage of viable tumor cells (p > 0.05); however, a trend towards less viable tumor cells in patients with high pazopanib concentrations in tumor tissue was observed in a categorical analysis. Possible explanations for the lack of correlation might be heterogeneity of the tumors and timing of the biopsy procedure.

The influence of delay in mononuclear cell isolation on acute myeloid leukemia phosphorylation profiles

Mass-spectrometry (MS) based phosphoproteomics is increasingly used to explore aberrant cellular signaling and kinase driver activity, aiming to improve kinase inhibitor (KI) treatment selection in malignancies. Phosphorylation is a dynamic, highly regulated post-translational modification that may be affected by variation in pre-analytical sample handling, hampering the translational value of phosphoproteomics-based analyses. Here, we investigate the effect of delay in mononuclear cell isolation on acute myeloid leukemia (AML) phosphorylation profiles. We performed MS on immuno-precipitated phosphotyrosine (pY)-containing peptides isolated from AML samples after seven pre-defined delays before sample processing (direct processing, thirty minutes, one hour, two hours, three hours, four hours and 24 h delay). Up to four hours, pY phosphoproteomics profiles show limited variation. However, in samples processed with a delay of 24 h, we observed significant change in these phosphorylation profiles, with differential phosphorylation of 22 pY phosphopeptides (p < 0.01). This includes increased phosphorylation of pY phosphopeptides of JNK and p38 kinases indicative of stress response activation. Based on these results, we conclude that processing of AML samples should be standardized at all times and should occur within four hours after sample collection. SIGNIFICANCE: Our study provides a practical time-frame in which fresh peripheral blood samples from acute myeloid patients should be processed for phosphoproteomics, in order to warrant correct interpretation of in vivo biology. We show that up to four hours of delayed processing after sample collection, pY phosphoproteomic profiles remain stable. Extended delays are associated with perturbation of phosphorylation profiles. After a delay of 24 h, JNK activation loop phosphorylation is markedly increased and may serve as a biomarker for delayed processing. These findings are relevant for biomedical acute myeloid leukemia research, as phosphoproteomic techniques are of particular interest to investigate aberrant kinase signaling in relation to disease emergence and kinase inhibitor response. With these data, we aim to contribute to reproducible research with meaningful outcomes.

Epithelial Transfer of the Tyrosine Kinase Inhibitors Erlotinib, Gefitinib, Afatinib, Crizotinib, Sorafenib, Sunitinib, and Dasatinib: Implications for Clinical Resistance

Simple Summary

Tyrosine kinase inhibitors (TKIs) specifically inhibit phosphorylation of signaling pathways of cancer cells, thereby inhibiting their growth. They are characterized by a poor solubility and high protein binding, leading to a large variability in gut uptake after oral administration and variation in the clinical efficacy. We used the CaCo2 gut epithelial model to characterize the gut absorption of 7 TKIs and observed a large variation in apical/basolateral (mimicking gut/blood) transfer, with 4 TKIs showing a negative and 3 a neutral transfer. A highly negative transfer may lead to pharmacokinetic resistance. Intracellular uptake of TKIs was high for sunitinib and crizotinib, intermediate for gefitinib, dasatinib and sorafenib, low for afatinib and not detectable for erlotinib. These properties may explain a high red blood cell to plasma ratio for most TKIs investigated. Although TKIs are poorly absorbed the latter property may compensate for this.

Abstract

Background: tyrosine kinase inhibitors (TKIs) inhibit phosphorylation of signaling proteins. TKIs often show large variations in the clinic due to poor pharmacology, possibly leading to resistance. We compared gut absorption of inhibitors of epidermal growth factor receptor (erlotinib, gefitinib, and afatinib), ALK-cMET (crizotinib), PDGFR/BCR-Abl (dasatinib), and multikinase inhibitors (sunitinib and sorafenib). In clinical samples, we measured the disposition of each compound within various blood compartments. Methods: we used an optimized CaCo2 gut epithelial model to characterize 20 µM TKI absorption. The apical/basolateral transfer is considered to represent the gut/blood transfer. Drugs were measured using LC-MS/MS. Results: sorafenib and sunitinib showed the highest apical/basolateral transfer (Papp 14.1 and 7.7 × 10⁻⁶ cm/s, respectively), followed by dasatinib (3.4), afatinib (1.5), gefitinib (0.38), erlotinib (0.13), and crizotinib (n.d.). However, the net absorptions for dasatinib, afatinib, crizotinib, and erlotinib were highly negative (efflux ratios >5) or neutral/negative, sorafenib (0.86), gefitinib (1.0), and sunitinib (1.6). A high negative absorption may result in resistance because of a poor exposure of tissues to the drug. Accumulation of the TKIs at the end of the transfer period (A->B) was not detectable for erlotinib, very low for afatinib 0.45 pmol/μg protein), followed by gefitinib (0.79), dasatinib (1.1), sorafenib (1.65), and crizotinib (2.11), being highest for sunitinib (11.9). A similar pattern was found for accumulation of these drugs in other colon cell lines, WiDr and HT29. In clinical samples, drugs accumulated consistently in red blood cells; blood to plasma ratios were all >3 (sorafenib) or over 30 for erlotinib. Conclusions: TKIs are consistently poorly absorbed, but accumulation in red blood cells seems to compensate for this.