Longitudinal Image Data for Outcome Modeling.

In oncology, medical imaging is crucial for diagnosis, treatment planning and therapy execution. Treatment responses can be complex and varied and are known to involve factors of treatment, patient characteristics and tumor microenvironment. Longitudinal image analysis is able to track temporal changes, aiding in disease monitoring, treatment evaluation, and outcome prediction. This allows for the enhancement of personalized medicine. However, analyzing longitudinal 2D and 3D images presents unique challenges, including image registration, reliable segmentation, dealing with variable imaging intervals, and sparse data. This review presents an overview of techniques and methodologies in longitudinal image analysis, with a primary focus on outcome modeling in radiation oncology.

Radiotherapy induces an increase in serum antioxidant capacity reflecting tumor response.

Background and purpose: Radiotherapy (RT) is a mainstay component of treatment for patients with head and neck squamous cell carcinoma (HNSCC), but responses vary. As RT relies upon oxidative damage, antioxidant expression in response to RT-induced reactive oxygen species (ROS) could compromise treatment response. We aimed to examine local and systemic antioxidant responses to increased RT-induced ROS in relation to treatment success. Materials and methods: Nuclear factor erythroid 2-related factor 2 (NRF2), the main antioxidant transcription factor, was immunofluorescently stained in FaDu cells and in tumor biopsies of patients with oral cavity/oropharynx HNSCC before and after five fractions of RT. Besides, total antioxidant capacity (TAC) was analyzed in HNSCC tumor cells in vitro and in serum of HNSCC patients before, during, and after RT. Results: Data revealed an increase in NRF2 expression and TAC in head and neck cancer cells in vitro over the course of 5 daily fractions of 2 Gy. In accordance, also in patients' tumors NRF2 expression increased, which was associated with increased serum TAC during RT. Increasing serum TAC was related to impaired local tumor control. Conclusion: Radiation induced NRF2 expression and upregulated TAC, which may compromise the effect of RT-induced ROS. Changes in serum TAC during RT could serve as a novel predictor of treatment outcome in HNSCC patients.Medical Ethics Review Committee (CMO) approval - CMO number: 2007/104.

Tamoxifen induces radioresistance through NRF2-mediated metabolic reprogramming in breast cancer.

BACKGROUND: Recently, we reported that tamoxifen-resistant (TAM-R) breast cancer cells are cross-resistant to irradiation. Here, we investigated the mechanisms associated with tamoxifen-induced radioresistance, aiming to prevent or reverse resistance and improve breast cancer treatment. METHODS: Wild-type ERα-positive MCF7 and ERα-negative MDA-MB-231 breast cancer cells and their TAM-R counterparts were analyzed for cellular metabolism using the Seahorse metabolic analyzer. Real-time ROS production, toxicity, and antioxidant capacity in response to H2O2, tamoxifen, and irradiation were determined. Tumor material from 28 breast cancer patients before and after short-term presurgical tamoxifen (ClinicalTrials.gov Identifier: NCT00738777, August 19, 2008) and cellular material was analyzed for NRF2 gene expression and immunohistochemistry. Re-sensitization of TAM-R cells to irradiation was established using pharmacological inhibition. RESULTS: TAM-R cells exhibited decreased oxygen consumption and increased glycolysis, suggesting mitochondrial dysfunction. However, this did not explain radioresistance, as cells without mitochondria (Rho-0) were actually more radiosensitive. Real-time measurement of ROS after tamoxifen and H2O2 exposure indicated lower ROS levels and toxicity in TAM-R cells. Consistently, higher antioxidant levels were found in TAM-R cells, providing protection from irradiation-induced ROS. NRF2, a main activator of the antioxidant response, was increased in TAM-R cells and in tumor tissue of patients treated with short-term presurgical tamoxifen. NRF2 inhibition re-sensitized TAM-R cells to irradiation. CONCLUSION: Mechanisms underlying tamoxifen-induced radioresistance are linked to cellular adaptations to persistently increased ROS levels, leading to cells with chronically upregulated antioxidant capacity and glycolysis. Pharmacological inhibition of antioxidant responses re-sensitizes breast cancer cells to irradiation.

Imageable Biomarkers for Radiotherapy Response.

Ideally, each patient with a malignancy who is eligible for radiation therapy should receive the most tumoricidal form of this this treatment with the lowest possible risk of toxicity. To overcome radiotherapy resistance, some patients would benefit from a more aggressive approach. This could be treatment intensification, for example by acceleration of the treatment to prevent the negative effects of accelerated tumor cell proliferation, or by boosting certain areas to specifically address intrinsic radioresistance, or a combination of radiotherapy with, for example, a hypoxic cell sensitizer or chemotherapy to reduce the radiotherapy resistance caused by hypoxia. For some patients, one of these approaches can be beneficial but for others could lead to unacceptable side effects. Therefore, it is highly desirable to make the selection upfront. The use of imageable biomarkers could be the key to a more patient-tailored treatment. Different biomarkers for hypoxia and proliferation that could be valuable for radiotherapy are discussed here, including their mechanism, the imaging procedure, quantification, and the value of the results.

Monitoring hypoxia and vasculature during bevacizumab treatment in a murine colorectal cancer model.

The purpose of this study was to assess the effect of bevacizumab on vasculature and hypoxia in a colorectal tumor model. Nude mice with subcutaneous LS174T tumors were treated with bevacizumab or saline. To assess tumor properties, separate groups of mice were imaged using (18) F-Fluoromisonidazole (FMISO) and (18) F-Fluorodeoxyglucose (FDG) positron emission tomography or magnetic resonance imaging before and 2, 6 and 10 days after the start of treatment. Tumors were harvested after imaging to determine hypoxia and vascular density immunohistochemically. The T2 * time increased significantly less in the bevacizumab group. FMISO uptake increased more over time in the control group. Vessel density significantly decreased in the bevacizumab-treated group. The Carbonic anhydrase 9 (CAIX) and glucose uptake transporter 1 (GLUT1) fractions were higher in bevacizumab-treated tumors. However, the hypoxic fraction showed no significant difference. Bevacizumab led to shorter T2 * times and higher GLUT1 and CAIX expression, suggesting an increase in hypoxia and a higher glycolytic rate. This could be a mechanism of resistance to bevacizumab. The increase in hypoxia, however, could not be demonstrated by pimonidazole/FMISO, possibly because distribution of these tracers is hampered by bevacizumab-induced effects on vascular permeability and perfusion.

18F-FLT PET changes during radiotherapy combined with cetuximab in head and neck squamous cell carcinoma patients.

AIM: Early treatment response of head and neck cancer to radiotherapy concomitant with cetuximab was monitored by repetitive PET imaging with the proliferation tracer 18F-FLT. PATIENTS, METHODS: Five head and neck cancer patients, treated with radiotherapy and concomitant cetuximab following cetuximab induction, received four 18F-FLT PET-CT scans before and during treatment. Changes in SUVpeak, SUVmean and CT- and PET-segmented gross tumour volumes were evaluated, as were correlations with immunohistochemical staining for Epidermal Growth Factor Receptor (EGFR) and Ki-67 (proliferation marker) in pre-treatment tumour biopsies. RESULTS: 18F-FLT PET measured tumor responses to the induction dose of cetuximab varied from 43% SUVpeak decrease to 47% increase. After start of radiotherapy 18F-FLT PET parameters decreased significantly in all patients. No associations were found between PET parameters and EGFR or Ki-67 expression levels. CONCLUSION: Proliferation of head and neck carcinomas shows a varying response to cetuximab induction, but consistently decreases after addition of radiotherapy.

Hypoxia and tumor metabolism in radiation oncology: targets visualized by positron emission tomography.

Due to the amazing leap of technology in radiation oncology in the past few years, cancer treatment will become more individualized. Molecular imaging with PET contributed to this with its many tracers available, each of them visualizing a specific feature of a tumor and its microenvironment revealing the biological characteristics of cancer. Hypoxia is of interest as hypoxic tumor cells are associated with lower disease control because of an increased resistance to cytotoxic treatment. This is especially the case for radiotherapy. Treatment adaptations overcoming the negative effect of hypoxia have shown promising results. Several hypoxia tracers are available of which [18F]FMISO is studied most extensively, however other tracers are studied as well and the search for highly specific and reproducible PET tracers is still ongoing. Wide experience has been gained with the use of [18F]FDG PET as it is used on a routine basis for diagnosing and staging of cancer. Although not a specific marker for hypoxia, increased metabolic rate reflects increased proliferation and glycolysis indicating increased treatment resistance. Molecular imaging by means of PET creates an opportunity to provide personalized care, with optimal disease control, minimal toxicity and best cost-effectiveness.

EGFR-inhibition enhances apoptosis in irradiated human head and neck xenograft tumors independent of effects on DNA repair.

Epidermal growth factor receptor (EGFR) inhibition using cetuximab improves the efficacy of radiotherapy in only a subgroup of head and neck squamous cell carcinoma (HNSCC) patients. Therefore, to improve patient selection a better understanding of tumor characteristics that affect treatment is necessary. Here, we investigated the effect of cetuximab on repair of radiation-induced DNA damage in a HNSCC xenograft model, which shows a synergistic effect to cetuximab and radiotherapy (SCCNij185) and a HNSCC model, which shows no additive effect of cetuximab to radiotherapy (SCCNij153). In both tumor models, clear increases were seen in the number of 53BP1 and Rad51 foci after irradiation. 53BP1 foci were present at comparable levels in hypoxic and normoxic tumor areas of the tumor xenografts, while the number of Rad51 foci was significantly higher in normoxic areas compared to hypoxic areas (P < 0.05). In both SCCNij185 and SCCNij153 xenografts an increased number of 53BP1 foci was observed in tumors treated with cetuximab and radiotherapy compared to radiotherapy alone. In SCCNij185 this increase was statistically significant in normoxic tumor areas (P = 0.04) and in SCCNij153 in both hypoxic and normoxic areas (P = 0.007 and P = 0.02, respectively). The number of Rad51 foci was not significantly different when cetuximab was added to radiotherapy compared to radiotherapy alone. Levels of pEGFR and pERK1/2 were decreased when cetuximab was added to radiotherapy in SCCNij185, but not in SCCNij153. Apoptosis was also only increased in SCCNij185 tumors at 4 days after cetuximab and radiotherapy treatment (P < 0.01). In conclusion, cetuximab inhibited DNA repair in both HNSCC models, but this effect was not predictive for the radiosensitizing effect of cetuximab in vivo. This lack of correlation may be related to differential effects of cetuximab and radiotherapy on ERK1/2 signaling and a decreased induction of apoptosis by cetuximab and radiotherapy in the resistant model.

Monitoring the effects of bevacizumab beyond progression in a murine colorectal cancer model: a functional imaging approach.

Clinical studies have shown that bevacizumab beyond progression to first line therapy is beneficial for overall survival in advanced stage colorectal cancer. We studied the utility of several functional imaging modalities to assess the efficacy of bevacizumab beyond progression (BBP). All BALB/c mice with s.c. LS174T xenografts were treated with capecitabine, oxaliplatin and bevacizumab combination therapy. Tumor volume was assessed using caliper measurements. Increase of 1.5 times the initial volume on two subsequent measurements, was considered progression. In half of the mice bevacizumab treatment was continued (n = 13) after progressive disease was established, while the others received saline injections (n = 12). Within 3 days after progression, multi-modal imaging was performed using FDG-PET, diffusion weighted imaging, T2* and dynamic contrast enhanced MRI. Measurements were repeated 7 and 10 days after the first measurements. Afterwards, tumors were analyzed for expression of carbonic anhydrase IX, glucose transporter 1, 9 F1 to stain the vasculature and Ki67 to assess proliferation. In the BBP group tumor growth after progression was reduced compared to the control group (p < 0.01). FDG-PET showed a trend towards lower FDG uptake in the BBP group (p = 0.08). DWI, T2* and DCE-MRI parameters were not significantly different between both groups. The immunohistochemical analyses showed higher CAIX-positive fraction (p < 0.01) and lower Ki67 expression (p = 0.06) in the BBP group. The relative vascular area was significantly lower in the BBP group (p = 0.03). GLUT-1 expression and vascular density did not significantly differ between both groups. Bevacizumab after progression resulted in significant changes in the tumor proliferation and microenvironment compared to discontinuation of bevacizumab. FDG-PET may be sensitive to BBP-induced effects.

Reproducibility and biological basis of in vivo T(2)* magnetic resonance imaging of liver metastasis of colorectal cancer.

In this study, the reproducibility of T2* MR imaging in colorectal liver metastases was assessed and T2* values were correlated with the expression of the hypoxia-related markers GLUT-1 and CA-IX as well as the relative vascular area, and the vessel density in resected tumors. The reproducibility of T2* was analyzed in 18 patients with in total 22 colorectal liver metastases using the Bland and Altman method for the 16th, 50th, and 84th percentile values. Immunohistochemical staining was performed on 17 resected tumors obtained from 16 patients. The median T2* of all liver metastases was 25.0 ± 5.6 ms vs. 23.0 ± 4.1 ms (median ± st.dev.) in normal liver. The coefficient of repeatability was 11.2 ms and the limits of agreement were -13.2 ms and 9.1 ms for median T2* values. On average, T2* showed fair reproducibility. No correlations between T2* values, hypoxia- and vascularity-related markers were observed.

SU-E-T-06: A Mathematical Explanation to Tumor's Response to Perfusion and Hypoxic Fraction after Radiation.

PURPOSE: To develop a dynamic model that explains oxygen dynamics between the microvascular perfusion and the hypoxic cell population inside a tumor. METHODS: Bussink et al (Radiat Res 153(4), p.398 (2000)) observed fast oxygen dynamics, faster than cell-death. Based on a simplified three-compartment-model: the microvasculature, well-oxygenated, and hypoxic tumor cell populations. We applied a first-order differential model for the tumor's transient response as a function of oxygen content within the blood vessels. The sink terms in our model for each compartment are fast changing parameters because radiation rapidly changes the oxygen consumption of the tumor cell in a time scale which is much faster than the population changes of the tumor. Transportation balance condition is also applied for each compartment. RESULTS: Our simulation results can explain the experimental data in Bussink et al's (Radiat Res 153(4), p.398 (2000)) paper. We provide an explanation for the relative complex behavior of the microvascular perfusion after radiation that emphasizes the role of dynamic metabolic changes in addition to population changes. CONCLUSIONS: A newly developed dynamic model leads our understanding to the interrelationship between microvascular oxygen content within the blood vessels and the hypoxia state of the tumor to a deeper level, which has the potential to provide the theoretical foundation for the patient' specific adaptive radiotherapy.

Quantification of tumour hypoxia. Functional histology and autoradiography.

Tumor cell hypoxia is considered one of the important causes for radiation resistance. The introduction of IMRT(intensity modulated radiotherapy) allows specific boosting of tumor subvolumes that may harbour these radioresistant tumour cells. PET imaging of these subvolumes can be incorporated into treatment planning. However, at this moment microenvironmental changes visualized and quantified by means of PET-imaging need to be validated by high-resolution microscopic techniques. This will allow interpretation of imaging techniques with intermediate resolution (such as PET/CT) in relation to complex cellular signaling in response to anti-cancer treatments.

Unusual case of bifocal leptomeningeal melanocytoma in the posterior fossa with seeding in the spinal canal.

A 26-year-old man presented with signs of raised intracranial pressure. CT and MRI of the head demonstrated two separate lesions in the posterior fossa. The radiological differential diagnoses included multiple meningiomas, schwannomas, neurofibromas and subependymomas. Both lesions were surgically resected. Histopathological examination revealed localisations of a leptomeningeal melanocytoma. Leptomeningeal melanocytoma is a rare tumour of the central nervous system. Generally, it has a good prognosis if radical resection can be performed. In cases of subtotal resection, adjuvant radiotherapy should be considered. Local recurrences are common. Less frequently, leptomeningeal metastases and, on rare occasions, distant metastases or progression to malignant melanoma have been described. We describe an unusual case with multiple localisations of melanocytoma in the posterior fossa and spinal canal, with the emphasis being on the radiological findings and diagnosis of this rare tumour. After surgery of the brain, this patient was irradiated on the craniospinal axis.

Angiogenesis, hypoxia and VEGF expression during tumour growth in a human xenograft tumour model.

Tumour growth and spread of tumour cells requires angiogenesis. Incipient angiogenesis is not induced by tumour cell hypoxia but probably by proangiogenic factors. During growth tumours depend on a further induction of vascular development for adequate oxygen and nutrient supply. If the oxygen supply is insufficient, the resulting hypoxia stimulates angiogenesis through upregulation of HIF-1 alpha and VEGF. VEGF upregulation is associated with a poor response to treatment and poor prognosis. The aim of the study was to analyze the interrelationship between hypoxia and angiogenesis during tumour growth. Therefore the tumour vasculature architecture and functional properties of the vessels were studied during subsequent phases of tumour growth in relation to hypoxia and VEGF-expression. Tumours from the human glioblastoma multiforme tumour line E106 were transplanted in athymic mice. Tumours were harvested at 2 days after transplantation and when tumours reached a mean size of 2, 4, 6, 8 and 10 mm. VEGF was present early in the onset of angiogenesis independent of HIF-1 alpha. During tumour growth VEGF increased from 0.94 to 7.27 ng/mg assessed by ELISA. However, there was increasing intratumoural heterogeneity in the architecture of the tumours, even in the largest tumours small well oxygenated areas were detected resembling the relatively well organized architecture of the smallest tumours. The observation that tumour vasculature develops in early phases under normoxic and at later phases under hypoxic conditions with the presence of both conditions in the larger tumours, suggested that anti-angiogenic therapy should be directed towards HIF-1 alpha dependent and HIF 1-alpha independent pathways.

Prophylactic G-CSF and antibiotics enable a significant dose-escalation of triplet-chemotherapy in non-small cell lung cancer.

In advanced non-small cell lung cancer (NSCLC) the clinical benefit of a platinum-based doublet is only modest, therefore, attenuated dosed three-drug combinations are investigated. We hypothesized that with adequate support a full dosed chemotherapy triplet is feasible. The study was designed as a dose finding study of paclitaxel in chemotherapy-naive patients. Paclitaxel was given as a 3-h infusion on day 1, followed by fixed doses of teniposide (or etoposide) 100mg/m(2) days 1, 3, 5 and cisplatin 80 mg/m(2) day 1 every 3 weeks. As myelotoxicity was expected to be the dose-limiting toxicity, prophylactic G-CSF and antibiotic support was evaluated. Indeed, paclitaxel 120 mg/m(2) resulted in dose-limiting neutropenia, despite G-CSF support. Teniposide/etoposide day 1, 3, 5 was less myelotoxic compared to day 1, 2, 3. G-CSF support allowed paclitaxel dose-escalation to 250 mg/m(2). The addition of prophylactic antibiotics enabled dose-escalation to 275 mg/m(2) without reaching MTD. In conclusion, G-CSF and antibiotics prophylaxis enables the delivery of a full dosed chemotherapy triplet in previously untreated NSCLC patients.

Modulation of hypoxia in murine liver metastases of colon carcinoma by nicotinamide and carbogen.

There is increasing evidence that modulation of tumor hypoxia may improve therapy outcome. However, most preclinical data are derived from subcutaneous rather than orthotopic tumor models. We investigated the effect of the hypoxia-modulating agents nicotinamide and carbogen on tumor hypoxia, tumor blood perfusion, and proliferative activity in liver metastases of the murine colon carcinoma line C26a. In untreated C26a liver metastases, we observed a considerable amount of hypoxia, similar to the amount in liver metastases of patients with colorectal cancer. Compared to untreated mice, we observed a significantly smaller hypoxic fraction in the liver metastases of mice treated with nicotinamide and carbogen breathing as single treatments or in combination. In the group of mice that underwent carbogen breathing, perfusion was significantly lower than in the untreated group, but the decrease was only marginal. The proliferative activity was similar in all groups. In C26a subcutaneous tumors, a similar effect on hypoxia has been observed that was, however, combined with a decrease in proliferative activity. The different effects of nicotinamide and carbogen on parameters of the tumor microenvironment in liver metastases and subcutaneous tumors suggest that the host tissue influences the mechanism by which nicotinamide and carbogen exert their effects. Since tumor hypoxia may be a clinical problem in colorectal liver metastases, our results open possibilities for further research on the effect of hypoxia modifiers on colorectal liver metastases to improve therapy outcome.

Carbonic anhydrase-9 expression levels and prognosis in human breast cancer: association with treatment outcome.

Here, we set out to assess CA9 expression levels by real-time quantitative RT-PCR in breast cancer tissue samples obtained from 253 patients, and correlated those with relapse-free (RFS) survival. The median follow-up time was 75 months (range 2-168 months). CA9 expression was mainly found in high-grade, steroid receptor negative cancer tissues. CA9 levels were not significantly associated with RFS (P=0.926, hazard ratio (HR)=0.99, 95% CI=0.80-1.22) in the total cohort of 253 patients. In multivariate analysis with other clinicopathological factors, CA9 (P=0.018, HR=0.77, 95% CI=0.62-0.96), the interaction of adjuvant chemotherapy with CA9 (P=0.009, HR=1.31, 95% CI=1.07-1.61) and the interaction of adjuvant endocrine therapy with CA9 (P<0.001, HR=1.41, 95% CI=1.20-1.66) all contributed significantly to the final model. These results indicate that patients with low CA9 levels benefit more from adjuvant treatment than do patients with high levels. Thus, the determination of CA9 levels could aid in the selection of patients who will not benefit from adjuvant therapy, and whose prognosis will more likely improve with other treatment modalities.