Ingfrid Haldorsen, MD, discusses the role of imaging in the diagnosis of patients with endometrial cancer.
Ingfrid Haldorsen, MD
Preoperative imaging continues to be a necessary step when planning for surgery and choosing adjuvant therapy for patients with endometrial cancer, explains Ingfrid Haldorsen, MD.
An observational study is investigating the efficacy of PET/MRI in detecting metastatic disease in patients with endometrial cancer (NCT02765698). The aim of this trial is to determine whether PET/MRI imaging will provide physicians with preoperative information about specific sites in the body where disease is present. In addition, this may allow for patients to undergo less extensive surgery.
Additionally, novel techniques are being explored. An ongoing diagnostic study is evaluating the accuracy of 3D ultrasounds in comparison with MRI and final histology to detect myometrial invasion of endometrial cancer (NCT03207061).
“Every patient should have access to imaging facilities in order to get optimal treatment,” said Haldorsen, professor and radiologist at the University of Bergen and Haukeland University Hospital.
In an interview with OncLive, Haldorsen shared more insight on the role of imaging in the diagnosis of patients with endometrial cancer.Haldorsen: Although endometrial cancer is formally a surgical and pathological stage, preoperative imaging is an important part of the primary diagnostic workup. It is important for surgical planning and [to determine the options for] adjuvant therapy.
For local staging of patients with deep myometrial and cervical stromal invasions, vaginal ultrasound and dynamic contrast-enhanced MRI are the imaging methods yielding the best diagnostic performance. Lymph node metastases and distant spread detected via PET/CT has a better diagnostic performance than other noninvasive imaging methods.
However, in recent years, there has been a dramatic advancement in the development of novel functional imaging techniques. These imaging techniques enable depiction of microstructural and microvascular features of the tissue that may be highly relevant for the observed tumor phenotype. Being able to fully exploit these advancements in endometrial cancer is important for patients. There are several challenges. One challenge in the routine diagnostic workup would be interobserver variation between radiologists and gynecologists when they are performing vaginal ultrasounds for preoperative staging. It is well known that interobserver variation exists, which somewhat limits the validity of these findings if they are going to guide a treatment in a valid way.
Also, we do not know how these imaging findings may be related to molecular targets that could be used for treatment selection. Combining the imaging data and the molecular data in the corresponding tumor tissue has to be done if it is going to make progress. We have all kinds of imaging data, as well as genetic data and molecular markers in the tumor tissue. These have to be analyzed using artificial intelligence techniques and machine-learning algorithms by computer scientists in collaboration with oncologists and radiologists if these issues are to be highlighted.
That is what we are aiming for in our group. We want to analyze this research collaboratively with other specialists who have differing expertise.I hope that imaging gives more certainty regarding the stage and prognosis prior to surgery so that the surgeon can be better prepared for what to do on the right patients. Furthermore, I hope that we will be able to provide more prognostic tools with which to guide adjuvant therapy after primary surgery. Ultimately, imaging is very important when it comes to detecting recurrent disease. All of these aspects of imaging should be kept in mind and every patient should have access to imaging facilities in order to get optimal treatment. I hope they understand that imaging in the primary diagnostic workup is important for the patient. Knowing the tumor extent and the aggressive features defined by preoperative imaging findings may help determine which patients should have the most aggressive treatment and which could have a more conservative treatment that is less invasive.
Ultimately, I hope that patients who do not need a very invasive or aggressive treatment regimen may refrain from having that therapy. We need to do more research about the clinical phenotype before treatment.
Other limitations and challenges are that we derive level biomarkers that could help in the selection of patients for different treatment regimens, but we don't know how these should be employed to guide treatment in a better way.