Preclinical Molecular Imaging Group

Dr Gabriela Kramer-Marek’s group uses cutting-edge biomedical imaging techniques to gain information about the way particular genes drive cancer progression.

Our group’s long-term goal is to develop specific biomarkers for detecting cancers and to evaluate these biomarkers in pre-clinical cancer models

Notwithstanding the remarkable clinical success of mAb-based treatment regimens, not all patients benefit from them. This can be attributed, at least in part, to the complexity of the tumour microenvironment and its considerable heterogeneity both in terms of the tumour and non-tumour cell components. These phenomena represent a huge challenge in identifying predictive biomarkers and stratifying patient populations for personalised therapy approaches.

Therefore, there is an urgent need to develop assays that will help in three ways:

  1. accurate patient selection
  2. understanding intrinsic resistance mechanisms or the emergence of acquired resistance following treatment initiation and
  3. choosing the most effective combination regimen in circumstances in which single-agent therapies are insufficiently effective.

Currently, the baseline expression level of antigens targeted by therapeutic mAbs can be analysed by methods such as: immunohistochemistry (IHC), flow cytometry, proteomics, or next-generation sequencing of tumour tissues acquired at diagnostic biopsy or intra-operatively. These techniques aid our understanding of how cancer cells adapt to treatment and become resistant, but such methods are inherently invasive, prone to sampling errors caused by inter- and intra-tumour heterogeneity of receptor expression within analysed biopsy specimens and do not lend themselves readily to repeated sampling.

Positron emission tomography (PET), using radiolabelled mAbs, antibody fragments or engineered protein scaffolds (immuno-PET), has the potential to acquire information non-invasively and can be highly complementary to analyses based on tissue acquisition. Accordingly, immuno-PET agents might accurately identify the presence and accessibility of the target and provide a rapid assessment of tumour response to a variety of treatments in a timely fashion (e.g. within 1-2 weeks of treatment initiation).

Furthermore, immuno-PET agents can provide information about the heterogeneity of both target expression and therapeutic response, which are increasingly recognised as key factors in treatment resistance. This especially relates to patients with advanced disease in whom target expression may vary from site to site and a biopsy of a single local or metastatic deposit may not accurately reflect the situation across the entire disease burden. Although introduction of immuno-PET into routine clinical practice may add complexity and increase costs, with appropriate use this imaging modality has the potential to identify patients likely to benefit from therapy and assess the efficacy of novel target-specific drugs.

Against this background, our research focuses on the development and characterisation of targeted-PET radiotracers, including protein-based theranostic agents that enable smart monitoring of immunotherapies and expand opportunities for personalised medicine approaches.

Early diagnosis and individualized therapy have been recognized as crucial for the improvement of cancer treatment outcome. While proper molecular characterization of individual tumour types facilitates choice of the right therapeutic strategies, early assessment of tumour response to therapy could allow the physicians to discontinue ineffective treatment and offer the patient a more promising alternative. Therefore, the role of molecular imaging in elucidating molecular pathways involved in cancer progression and the ability to select the most effective therapy based on the unique biologic characteristics of the patient and the molecular properties of a tumour are undoubtedly of paramount importance.

The mission of this group is to investigate innovative imaging probes and apply them to novel orthotopic or metastatic models that are target driven, to gain information of the way particular oncogenes drive cancer progression through signalling pathways that can be imaged in vivo and, correlate it with target level ex vivo. Such an approach enables non-invasive assessment of biochemical target levels, target modulation and provides opportunities to optimize the drug dosing for maximum therapeutic effect, which leads to the development of better strategies for the more precise delivery of medicine.

The long term goal of our research is to develop specific imaging cancer biomarkers, especially for positron emission tomography (PET) as well as optical imaging and, evaluate these biomarkers in pre-clinical cancer models. Significant efforts are directed towards validating biomarkers for early prediction of treatment response, with the focus on new targeted therapies (such as inhibition of cell signalling pathways).

Our initial portfolio of imaging agents include radiolabelled affibody molecules, TK inhibitors and, conventional tracers that monitor universal markers of tumour physiology.

We are actively supported by other groups from the Division of Radiotherapy and Imaging as well as the Division of Cancer Therapeutics. Moreover, our close association with The Royal Marsden NHS Foundation Trust enables rapid translation of our research to early clinical studies and ensures a fast transition of know-how from the research laboratory to the patient bedside.

Dr Gabriela Kramer-Marek

Group Leader:

Preclinical Molecular Imaging Gabriela Kramer-Marek

Dr Gabriela Kramer-Marek is investigating new ways of molecular imaging in order to predict an individual patient’s response to treatment. Before moving to the ICR, she developed a new approach for non-invasive assessment of HER2 expression in breast cancer.

See profile

Researchers in this group

.

Phone: 020 3437 6376

Email: [email protected]

Location: Sutton

See profile

.

Phone: +44 20 3437 6785

Email: [email protected]

Location: Sutton

See profile

.

Phone: 020 3437 6356

Email: [email protected]

See profile

.

Phone: +44 20 3437 6857

Email: [email protected]

Location: Sutton

See profile

.

Phone: 020 3437 4549

Email: [email protected]

Location: Sutton

See profile

Dr Gabriela Kramer-Marek's group have written 63 publications

Most recent new publication 2/2025

See all their publications

Recent discoveries from this group

22/08/23

  A PET scan of a mouse with a brain tumour

Image: A PET scan showing glioblastoma in a mouse model.
Credit: Preclinical Molecular Imaging team at the ICR using the Albira (PET/SPECT/CT) Bruker system

A new form of screening may improve survival rates among people with a fast-growing type of brain tumour by helping identify those most likely to benefit from certain treatments.

Innovative preclinical research in mice models has shown that a molecular imaging technique can reveal the presence of a protein called PD-L1 in glioblastomas – the most common type of cancerous brain tumour in adults.

Alongside other measures, tests to detect high levels of PD-L1 could help direct treatment decisions, potentially leading to better patient outcomes.

Assessing PD-L1 expression levels

Currently, scientists assess PD-L1 expression levels by carrying out immunohistochemistry on samples of tissue taken from the patient during surgery, which is the first-line treatment for glioblastoma. However, this technique is subject to human error and is not standardised globally for these patients or this particular tumour. It can also be difficult to quantify the results.

Researchers at The Institute of Cancer Research have now shown that a non-invasive imaging technique called immuno-positron emission tomography (immuno-PET) could be a better approach.

The research has been published in the journal Cancers. It was largely funded by the ICR – which is both a research institute and a charity – and partially funded by the Cancer Research UK Convergence Science Centre at the ICR, Imperial College London and the National Science Centre in Poland.

Immunotherapy may improve the treatment landscape in glioblastoma

Glioblastoma starts as a growth of cells in the brain. It grows quickly and typically spreads within the brain, making it very difficult to treat effectively. No cure is yet available, and patients who initially respond to treatment tend to experience relapse. The average survival time is just 12–18 months, with only 5 per cent of patients surviving more than five years.

In recent years, immunotherapy has shown potential as a treatment for glioblastoma. In particular, researchers have been testing drugs called immune checkpoint inhibitors, which prevent other proteins from dampening the body’s immune response. The results to date have been mixed, suggesting that the treatment is only likely to be effective for a subset of patients.

Creating a novel radiotracer

The ICR’s team successfully used NOTA-maleinide to link ZPD-L1 affibody molecules to fluorine-18 and gallium-68 radionuclides. Affibodies are small proteins created to bind strongly to target proteins – in this case, PD-L1. This procedure created 18F-AIF-NOTA-ZPD-L1 and 68Ga-NOTA-ZPD-L1, which, with high specificity, recognise PD-L1 on tumour cells and in their microenvironment.

The team chose to use affibodies rather than antibodies because their much smaller size means that they clear the body far more quickly, minimising the radiation dose for patients and preventing delays to surgery. Using affibodies also makes it possible to get high-quality images just one hour after injection. In comparison, when antibodies are used, the images are usually only retrieved after 48 hours.

Trialling the new approach

The researchers demonstrated that these radiolabelled affibodies could be used to assess the expression level of PD-L1 in tumours in mice. PET scans showed that although there was some uptake of the radiotracer in healthy tissue, the brain tumours were clearly visualised with high tumour-to-background contrast.

Then, the researchers looked into 36 samples from people with newly diagnosed glioblastoma. They noted PD-L1-positive membrane staining in 39 per cent of the samples. A separate analysis of 161 human glioblastoma samples confirmed that tumours with a mesenchymal signature, which is linked to a better response to immune checkpoint inhibitors, had a significantly elevated expression level of PD-L1 compared with other glioblastoma subtypes. This supports the thinking that healthcare professionals could use immuno-PET to identify the patients most likely to benefit from immune checkpoint inhibitors.

‘Really exciting’

The researchers hope that this work will lead to better outcomes for the 30–49 per cent of patients with the mesenchymal subtype of glioblastoma. They are now working on a clinical trial in Poland that builds on the foundations laid by this preclinical research and expect to present data from that trial in the near future.

Dr Gabriela Kramer-Marek, Group Leader in Preclinical Molecular Imaging at the ICR, said:

“It has been really exciting to see the journey from lab to clinic. We are currently running a clinical trial in people, which was only possible because of this promising preclinical work. The trial was the first ever to use immuno-PET to evaluate PD-L1 in people with primary glioblastoma, and we hope to see images that clearly show the presence of PD-L1 in these brain tumours.

“The treatment for glioblastoma has not changed for decades. Although we still do not have a cure, I believe that this new screening approach could definitely change patient outcomes.”

See all discoveries