Validation and qualification of a multiplex hypoxia biomarker for radiotherapy individualisation
Lead Research Organisation:
University of Manchester
Department Name: Medical and Human Sciences
Abstract
What is the background of this research?
Approximately 300,000 patients develop cancer in the UK each year, and around 50% will undergo radiotherapy at some point during treatment. Hypoxia (meaning ?low oxygen?) in a tumour is associated with resistance to radiotherapy. If doctors could identify whether a tumour was hypoxic, it would help them improve treatment to give the best chance of cure for a patient.
What is the research trying to achieve?
Previous research identified a group of 99 genes that are linked with hypoxia in head and neck tumours. More recently this group has been refined to a subset that comprises just 30 genes. This research aims to confirm whether the reduced 30-gene group can be used in the clinic as a test for hypoxia.
To do this, we will be using tumour samples collected from patients who chose to take part in a clinical trial. We know there is a special pattern of activity in these 30 genes that reflects oxygen levels in a tumour. We are investigating different ways of identifying this pattern to find the one which is most reliable and reproducible, and therefore most useful in the clinic.
We also want to investigate whether our pattern of 30 genes can identify and predict hypoxia in other cancers e.g. bladder. We will investigate whether a different 30-gene group is needed for each cancer type.
Why is this important?
If we are able to develop a reliable test for hypoxia, we could try it out in a clinical trial. If successful it could be introduced into clinical practice. This would mean patients would receive the treatment that is best for them, which would improve outcome. Given the large number of cancer patients who undergo radiotherapy each year, this would have a significant benefit on cancer survival in the UK.
How and where is the research to be conducted?
This study builds on long-standing collaborations between scientists and doctors from across the UK and EU. It will be led by Dr. Catharine West at the University of Manchester, who has many years experience of measuring tumour hypoxia in patients. This study will use tumour samples collected from patients who gave full and informed consent for their tissue to be used in future ethically approved research.
Approximately 300,000 patients develop cancer in the UK each year, and around 50% will undergo radiotherapy at some point during treatment. Hypoxia (meaning ?low oxygen?) in a tumour is associated with resistance to radiotherapy. If doctors could identify whether a tumour was hypoxic, it would help them improve treatment to give the best chance of cure for a patient.
What is the research trying to achieve?
Previous research identified a group of 99 genes that are linked with hypoxia in head and neck tumours. More recently this group has been refined to a subset that comprises just 30 genes. This research aims to confirm whether the reduced 30-gene group can be used in the clinic as a test for hypoxia.
To do this, we will be using tumour samples collected from patients who chose to take part in a clinical trial. We know there is a special pattern of activity in these 30 genes that reflects oxygen levels in a tumour. We are investigating different ways of identifying this pattern to find the one which is most reliable and reproducible, and therefore most useful in the clinic.
We also want to investigate whether our pattern of 30 genes can identify and predict hypoxia in other cancers e.g. bladder. We will investigate whether a different 30-gene group is needed for each cancer type.
Why is this important?
If we are able to develop a reliable test for hypoxia, we could try it out in a clinical trial. If successful it could be introduced into clinical practice. This would mean patients would receive the treatment that is best for them, which would improve outcome. Given the large number of cancer patients who undergo radiotherapy each year, this would have a significant benefit on cancer survival in the UK.
How and where is the research to be conducted?
This study builds on long-standing collaborations between scientists and doctors from across the UK and EU. It will be led by Dr. Catharine West at the University of Manchester, who has many years experience of measuring tumour hypoxia in patients. This study will use tumour samples collected from patients who gave full and informed consent for their tissue to be used in future ethically approved research.
Technical Summary
Background: Approximately 300,000 patients develop cancer in the UK each year, and ~50% will undergo radiotherapy during their treatment. Tumour hypoxia drives cancer treatment resistance and reduces the efficacy of radiotherapy. There is evidence hypoxia-modification strategies improve outcome in head and neck (H&N) cancer patients and that those with hypoxic vs oxygenated tumours are most likely to benefit.
Microarray gene expression profiling of 59 H&N tumours produced a hypoxia-associated 99-gene signature. The signature was prognostic in a publicly available H&N cancer dataset outperforming the original intrinsic classifier, and was an independent prognostic factor in a widely studied breast cancer series. A reduced and refined 30-gene hypoxia meta-signature was developed, which has similar or greater prognostic significance than the 99-gene signature in publicly available datasets.
Research objective: To validate and qualify the 30-gene hypoxia biomarker taking it to the final stage of qualification i.e. prospective evaluation in a clinical trial.
Methodology: Working to GCLP, RNA will be extracted from fresh tumour samples and archival FFPE blocks. The clinical samples/datasets available are: prospectively-collected fresh and FFPE H&N samples with parallel assessment of hypoxia (pimonidazole); a retrospective series of FFPE cervix samples with parallel assessment of hypoxia (pO2, pimonidazole, CA9, HIF-1alpha) and complete radiotherapy outcome data; bladder samples from the UK BCON trial; H&N samples from the Dutch ARCON trial; H&N samples from the Danish DAHANCA5 trial.
The 30-gene hypoxia biomarker will be validated by assessing the repeatability and reproducibility of a PCR card approach (prospective H&N samples). The biomarker will be compared with other methods for assessing hypoxia (pO2, pimonidazole, expression of hypoxia related markers (CA9, HIF-1alpha; prospective H&N and cervix). Whole genome arrays, where signatures associated with multiple phenotypes can be assessed, will be compared with PCR cards, which have potential for greater sensitivity (cervix, bladder). The application of the biomarkers will be investigated by studying its prognostic significance in multiple cancers (cervix, bladder, H&N). A generic, tumour-type independent hypoxia meta-signature will be compared with tumour-specific signatures. Samples from 3 trials involving randomisation to radiotherapy with/without hypoxia modification will be used to determine ability of the hypoxia biomarker to predict benefit from hypoxia modification (i.e. biomarker qualification): BCON, ARCON and DAHANCA5.
Exploitation: The introduction in clinical practice of tumour hypoxia testing and the biological individualisation of radiotherapy. Given the large number of cancer patients who undergo radiotherapy each year, this would have a significant benefit on cancer survival in the UK.
Microarray gene expression profiling of 59 H&N tumours produced a hypoxia-associated 99-gene signature. The signature was prognostic in a publicly available H&N cancer dataset outperforming the original intrinsic classifier, and was an independent prognostic factor in a widely studied breast cancer series. A reduced and refined 30-gene hypoxia meta-signature was developed, which has similar or greater prognostic significance than the 99-gene signature in publicly available datasets.
Research objective: To validate and qualify the 30-gene hypoxia biomarker taking it to the final stage of qualification i.e. prospective evaluation in a clinical trial.
Methodology: Working to GCLP, RNA will be extracted from fresh tumour samples and archival FFPE blocks. The clinical samples/datasets available are: prospectively-collected fresh and FFPE H&N samples with parallel assessment of hypoxia (pimonidazole); a retrospective series of FFPE cervix samples with parallel assessment of hypoxia (pO2, pimonidazole, CA9, HIF-1alpha) and complete radiotherapy outcome data; bladder samples from the UK BCON trial; H&N samples from the Dutch ARCON trial; H&N samples from the Danish DAHANCA5 trial.
The 30-gene hypoxia biomarker will be validated by assessing the repeatability and reproducibility of a PCR card approach (prospective H&N samples). The biomarker will be compared with other methods for assessing hypoxia (pO2, pimonidazole, expression of hypoxia related markers (CA9, HIF-1alpha; prospective H&N and cervix). Whole genome arrays, where signatures associated with multiple phenotypes can be assessed, will be compared with PCR cards, which have potential for greater sensitivity (cervix, bladder). The application of the biomarkers will be investigated by studying its prognostic significance in multiple cancers (cervix, bladder, H&N). A generic, tumour-type independent hypoxia meta-signature will be compared with tumour-specific signatures. Samples from 3 trials involving randomisation to radiotherapy with/without hypoxia modification will be used to determine ability of the hypoxia biomarker to predict benefit from hypoxia modification (i.e. biomarker qualification): BCON, ARCON and DAHANCA5.
Exploitation: The introduction in clinical practice of tumour hypoxia testing and the biological individualisation of radiotherapy. Given the large number of cancer patients who undergo radiotherapy each year, this would have a significant benefit on cancer survival in the UK.
