Cytotoxicity determination without photochemical artifacts

Cytokine Profiles of Head and Neck Squamous Cell Carcinoma Undergoing Dual Immunotherapy With Cetuximab and Pembrolizumab Identify Interferon Gamma-Induced Protein 10 as Novel Biomarker

Background

Pembrolizumab and cetuximab are antibodies under investigation in head and neck squamous cell carcinoma (HNSCC) either as single agents or combined with cisplatin and other chemotherapeutic drugs, e.g., 5-fluorouracil and/or docetaxel. However, also the combination of both antibodies may have potential in recurrent/metastatic (R/M) HNSCC, in particular in cisplatin-resistant or -refractory cases or patients with comorbid disease, e.g. patients with impaired renal function.

Methods

To clarify potential benefit that may result from such combination, we used the FLAVINO assay, a short-time ex vivo assay to compare responsiveness of HNSCC to pembrolizumab, cetuximab and both combined regarding colony formation of epithelial cells of biopsy-derived tumor samples and their cytokine production within three days either without or with stimulation with 10 ng/mL interferon gamma (IFN-γ). Vascular endothelial growth factor A (VEGF), monocyte chemoattractant protein 1 (MCP-1 or CCL2), interleukin 6 (IL-6), IL-8, IFN-γ, and interferon gamma-induced protein 10 (IP-10 or CXCL10) in supernatants were measured by ELISA.

Results

We detected huge heterogeneity in response to cetuximab, pembrolizumab and both combined with and without IFN-γ stimulation. Moreover, we detected a link between IFN-γ induced IP-10 release and improved outcome in those HNSCC patients who were capable to respond to IFN-γ and pembrolizumab, cetuximab and both combined with a further increase in IP-10 production. We derived an “IP-10 score” that independent from clinical characteristics of HNSCC patients and therapy regimens applied was able to predict their outcome.

Conclusions

The heterogeneity in the ex vivo response of cetuximab, pembrolizumab and both combined with and without IFN-γ stimulation identifies subgroups of HNSCC patients with deviating OS.

Generation of Stable cisPt Resistant Lung Adenocarcinoma Cells

Platinum compounds represent the backbone of combined chemotherapy protocols for advanced lung cancer. The mechanisms responsible for its frequent primary or acquired resistance to cisplatin (cisPt)-based chemotherapy remains enigmatic. The availability of two cell lines of the same origin, one resistant and the other sensitive, will facilitate research to reveal the mechanism of resistance formation. Lung adenocarcinoma cells, A240286S (A24), were cultivated in increasing cisPt concentrations over a prolonged time. After a significant increase in IC₅₀ was measured, cultivation of the cells was continued in absence of cisPt and IC₅₀s determined over a long period (>7 months). As a result, a cell line with lasting, high-level cisPt resistance, designated (D-)A24cisPt8.0, was obtained. The cells were cross-resistant to oxaliplatin and to pemetrexed at a low level. Previous publications have claimed that Leucine-rich repeat-containing protein 8 (LRRC8A and LRRC8D) of the volume-regulated anion channels (VRACs) affect cellular resistance to cisPt. Even though cisPt decreased LRRC8D expression levels, we showed by knockdown and overexpression experiments with LRRC8A and D that these proteins do not govern the observed cisPt resistance. The tumor cell sublines described here provide a powerful model to study the mechanisms of resistance to cisPt in lung cancer cells and beyond.

Arginase inhibitor, Nω-hydroxy-L-norarginine, spontaneously releases biologically active NO-like molecule: Limitations for research applications

There has been a renewed interest in the enzyme arginase for its role in various physiological and pathological processes that go beyond the urea cycle. One such role ascribed to arginase has been that of regulating nitric oxide (NO) production by a substrate (l-arginine) competition between arginase and nitric oxide synthase (NOS). Several arginase inhibitors have been developed to investigate the biological roles of arginase, of which N^ω-hydroxy-l-norarginine (nor-NOHA) is commercially available and is used widely from cell culture models to clinical investigations in humans. Despite the prevalence of nor-NOHA to investigate the substrate competition between arginase and NOS, little is known regarding interferences that nor-NOHA could have on common methods to assess NO production. Therefore, we investigated if nor-NOHA has unintended consequences on common NO assessment methods. We show that nor-NOHA spontaneously releases biologically active NO-like molecule in cell culture media by reacting with riboflavin. This NO-like molecule is indistinguishable from an NO donor (NOR-3) using common methods to assess NO. Besides riboflavin, nor-NOHA spontaneously reacts with H₂O₂ to diminish H₂O₂ content and produce NO-like molecule in the process. Our investigation provides detailed evidence on unintended artefacts related to nor-NOHA that can limit its use in cell culture, as well as some ex vivo and in vivo models. Future studies on arginase should take into consideration the limitations presented here when using nor-NOHA as a research tool, not only in investigations related to arginase and NOS competition, but also for investigating other biological roles of arginase.

Chemoresponse Assay in Head and Neck Cancer Patients: A Three-Year Follow Up

INTRODUCTION

The majority of patients with advanced head and neck cancer receiving chemotherapy show partial response or frank resistance. Therefore, assessing the individuals' tumour reactivity to the eligible chemotherapeutic compounds carries the potential of personalizing the patient treatment and minimizing ineffective regimens which lead to excess toxicity and cost, treatment delays and possibly causing the tumour to be cross resistant to additional drugs.

AIM

To determine the effectiveness of a phenotypic chemoresponse assay in predicting response to chemotherapy in a retrospective series of head and neck cancer patients whose tumour specimens had been tested with ChemoFx assay (Precision Theraputic Inc.).

MATERIALS AND METHODS

Twenty-two tumour specimens were submitted to Precision Theraputics Inc. for chemoresponse testing, all of which have been histologically confirmed as squamous cell carcinoma of the head and neck. Selection of treatment was at the discretion of the treating physician and the results of the assay were not used to determine the therapy. A portion of the patients' solid tumour was established in primary culture, then exposed to increasing doses of different chemotherapeutic agents. The resultant cell counts in the treated wells were used to indicate the tumours' response to the agent and based on the dose response score curve, the test was scored as "responsive," "intermediate response," or "non-responsive."

RESULTS

Of the 22 tumour samples submitted, 16 (72.7%) showed adequate cell yield in cultures and subsequently underwent in vitro chemoresponse assays and are reported in this study. Of the 16 cases reviewed, 5 were excluded due to inadequate follow up. A predictable response assay was either a good response to chemotherapy in patients whose tumour specimens showed sensitivity to the chemotherapeutic agents or failure in patients whose tumours showed either intermediate response or non responsiveness to the chemotherapeutic agent/agents. Of the 11 patients reported in this study, nine showed a predictable chemoresponse assay (81.8% predictability of effective treatment). Three patients had a predictable good response and six who failed their chemotherapy regimen within six months of treatment and their chemoresponse assay showed an inadequate response to the chemotherapeutic agents they were treated with. At three years follow up, all patients who had a predictable poor response succumbed to their disease except one, whose test showed intermediate response.

CONCLUSION

While the current report has its limitation, we conclude, based on our findings, that chemoresponse assays may be useful adjuncts in the guiding the selection of chemotherapeutic agents in patients with head and neck cancer.

Assay-based response evaluation in head and neck oncology: requirements for better decision making

This article gives an overview on different current strategies of assay-based response evaluation in head and neck squamous cell carcinomas (HNSCC) and critically summarizes their role and needs for future clinical evaluation. Due to a growing amount of data of phase III clinical trials of multimodality treatment options for HNSCC, treatment planning in regard to optimal outcome is becoming an interdisciplinary challenge. New concepts such as induction chemotherapy with bi- or ternary combinations of chemotherapeutics, integration of targeted therapies, concurrent and sequential chemoradiation concepts, and multimodality-based organ preservation strategies strongly compete with traditional definitive surgical procedures. Moreover, the outcome is difficult to predict due to heterogeneity of a tumor's response, impaired late functional outcome, and increased late toxicity if simultaneously applied to radiation. Retrospectively looking at non-responders with tumors classified as resectable, primary surgery is very likely to have achieved better results, since chemoradiation causes a high degree of early and late toxicities leading to extremely complicated terms and conditions in surgery following current multimodal therapeutic strategies. Unfortunately, predictive information on response characteristics of a given tumor before starting the therapy is not available in daily routine, although heterogeneity in response of a given tumor entity to treatments has been known for decades. Therefore, current therapy strategies for HNSCC still have to ignore this fact, creating an urgent need for the development of proper predictive assays. There are interesting clinical observations showing that response on induction chemotherapy may predict the outcome after radiotherapy. Some trials use this empiric phenomenon to pre-select non-responders for primary surgical treatment avoiding severe salvage complications after failure of complete chemoradiation treatment. Moving one step further, recent literature and our own investigations implicate that response evaluation of the individual patient's HNSCC in a suitable ex vivo assay just before starting the treatment is mature for clinical research. To this end, essential needs and hints are addressed and discussed.

Heterogeneity of epithelial and stromal cells of head and neck squamous cell carcinomas in ex vivo chemoresponse

BACKGROUND

Valid prediction of the effectiveness of chemotherapeutic agents in individual head and neck squamous cell carcinoma (HNSCC) is desirable and might be achieved using ex vivo assays.

METHODS

Three biopsies from each of 15 HNSCC were taken, minced and digested by collagenase. The digested HNSCC was added to serial dilutions of either cisplatin (CIS) or docetaxel (DTX), which were prepared under flavin-protecting conditions in ECM-coated microtiterplates. After 72-h incubation, cultures were methanol-fixed and Giemsa-stained. The cutoff concentration (COC; concentration completely suppressing colony formation) for epithelial cells (EC) and stromal cells (SC) was evaluated.

RESULTS

12/15 HNSCC (80%) were evaluable. Despite significant correlation of COC of CIS in respect of colony formation of EC or SC, no significant differences in response of individual HNSCC specimens were found in the t test for paired samples (p > 0.16). The same applied to DTX. However, EC and SC showed heterogeneity in chemoresponses leading to COC variability of more than one titration step in 44.1% (CIS) and 20% of HNSCC (DTX). No significant correlation between the COC of both cell populations was found in HNSCC specimens.

CONCLUSIONS

The ex vivo chemoresponse of EC and SC of HNSCC must be analyzed separately.

Modeling the functional heterogeneity of leukemia stem cells: role of STAT5 in leukemia stem cell self-renewal

Although the cancer stem cell (CSC) concept implies that CSCs are rare, recent reports suggest that CSCs may be frequent in some cancers. We hypothesized that the proportion of leukemia stem cells would vary as a function of the number of dysregulated pathways. Constitutive expression of MN1 served as a 1-oncogene model, and coexpression of MN1 and a HOX gene served as a 2-oncogene model. Leukemia-initiating cell (LIC) number and in vitro expansion potential of LICs were functionally assessed by limiting dilution analyses. LIC expansion potential was 132-fold increased in the 2- compared with the 1-oncogene model, although phenotypically, both leukemias were similar. The 2-oncogene model was characterized by granulocyte-macrophage colony-stimulating factor (GM-CSF) hypersensitivity and activated STAT/ERK signaling. GM-CSF hypersensitivity of the 2-oncogene model (MN1/HOXA9) was lost in Stat5b(-/-) cells, and the LIC expansion potential was reduced by 86- and 28-fold in Stat5b(-/-) and Stat1(-/-) cells, respectively. Interestingly, in 201 acute myeloid leukemia (AML) patients, coexpression of MN1 and HOXA9 was restricted to patients with the poorest prognosis and was associated with highly active STAT signaling. Our data demonstrate the functional heterogeneity of LICs and show that STAT signaling is critical for leukemia stem cell self-renewal in MN1- and HOXA9-expressing leukemias.

A systematic review of platinum and taxane resistance from bench to clinic: An inverse relationship

We undertook a systematic review of the pre-clinical and clinical literature for studies investigating the relationship between platinum and taxane resistance. Medline was searched for (1) cell models of acquired drug resistance reporting platinum and taxane sensitivities and (2) clinical trials of platinum or taxane salvage therapy in ovarian cancer. One hundred and thirty-seven models of acquired drug resistance were identified. 68.1% of cisplatin-resistant cells were sensitive to paclitaxel and 66.7% of paclitaxel-resistant cells were sensitive to cisplatin. A similar inverse pattern was observed for cisplatin vs. docetaxel, carboplatin vs. paclitaxel and carboplatin vs. docetaxel. These associations were independent of cancer type, agents used to develop resistance and reported mechanisms of resistance. Sixty-five eligible clinical trials of paclitaxel-based salvage after platinum therapy were identified. Studies of single agent paclitaxel in platinum-resistant ovarian cancer where patients had previously recieved paclitaxel had a pooled response rate of 35.3%, n=232, compared to 22% in paclitaxel naïve patients n=1918 (p<0.01, Chi-squared). Suggesting that pre-treatment with paclitaxel may improve the response of salvage paclitaxel therapy. The response rate to paclitaxel/platinum combination regimens in platinum-sensitive ovarian cancer was 79.5%, n=88 compared to 49.4%, n=85 for paclitaxel combined with other agents (p<0.001, Chi-squared), suggesting a positive interaction between taxanes and platinum. Therefore, the inverse relationship between platinum and taxanes resistance seen in cell models is mirrored in the clinical response to these agents in ovarian cancer. An understanding of the cellular and molecular mechanisms responsible would be valuable in predicting response to salvage chemotherapy and may identify new therapeutic targets.

MN1 overexpression induces acute myeloid leukemia in mice and predicts ATRA resistance in patients with AML

Overexpression of wild-type MN1 is a negative prognostic factor in patients with acute myeloid leukemia (AML) with normal cytogenetics. We evaluated whether MN1 plays a functional role in leukemogenesis. We demonstrate using retroviral gene transfer and bone marrow (BM) transplantation that MN1 overexpression rapidly induces lethal AML in mice. Insertional mutagenesis and chromosomal instability were ruled out as secondary aberrations. MN1 increased resistance to all-trans retinoic acid (ATRA)–induced cell-cycle arrest and differentiation by more than 3000-fold in vitro. The differentiation block could be released by fusion of a transcriptional activator (VP16) to MN1 without affecting the ability to immortalize BM cells, suggesting that MN1 blocks differentiation by transcriptional repression. We then evaluated whether MN1 expression levels in patients with AML (excluding M3-AML) correlated with resistance to ATRA treatment in elderly patients uniformly treated within treatment protocol AMLHD98-B. Strikingly, patients with low MN1 expression who received ATRA had a significantly prolonged event-free (P = .008) and overall (P = .04) survival compared with patients with either low MN1 expression and no ATRA, or high MN1 expression with or without ATRA. MN1 is a unique oncogene in hematopoiesis that both promotes proliferation/self-renewal and blocks differentiation, and may become useful as a predictive marker in AML treatment.