Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Diagnosis

Cerebral metastases are the most common brain tumor in clinical practice since up to 40 % of all patients with a history of cancer will develop metastatic lesions in the brain. Therefore, all cancer patients should undergo an imaging investigation of the brain in regular time intervals which must not exceed, in most cases, 6 months. Additionally, all cancer patients with new neurological symptoms or progressive neurological deficits should promptly perform an imaging of the relevant parts of the central nervous system.

The principal modality for this is MRI. Although CT is cheaper, faster, and easier to perform, it has, in general, a lower screening, diagnostic, and treatment planning value compared to the MRI unless the suspected lesion affects or destroys bone structures- this is the case with most spine metastases- or has prominent calcifications. Iodine contrast enhanced CT of the brain can detect fast growing tumors such as high grade gliomas and metastases but still has the disadvantage of inferior tissue differentiation and anatomical definition compared to MRI, especially in the posterior fossa. Most brain metastases appear in the MRI as well circumscribed, enhancing lesions located at the gray-white matter junction, causing intense edema. More than 70 % of all patients with brain metastases have multiple lesions. Differential diagnoses include astrocytoma (especially high grade glioma), abscess, and non-specific inflammatory reaction.

Distinguishing a high grade primary brain neoplasm from a solitary metastatic lesion is not always an easy task but physiology-based MRI sequences can provide a certain aid. Low apparent diffusion coefficients (ADCs) measured in diffusion-weighted MRI (DWI) are a sign of high tumor cellularity and primary neoplasms infiltrating the surrounding brain parenchyma tend to have lower values in the peritumoral area compared to cerebral metastases. Perfusion MRI (PWI) shows also lower relative cerebral blood volume (rCBV) measurements outside the enhancing portion of the lesion in secondary neoplasms. MRS interrogation in the same areas that shows choline/NAA ratios greater than 1 has an excellent sensitivity for primary brain tumors [1].

Primary brain tumors are most often of glial origin. Low grade gliomas (LGG) are typically solitary lesions, hypointense on T1 weighted MRI images and a diffuse growth pattern on T2 weighted images. Thirty percent of LGGs show contrast enhancement. High grade gliomas (grades III–IV) are complex or ring enhancing lesions and frequently present with a central cyst or necrosis. DWI and MRS have not shown great clinical value in tumor grading of gliomas, but increased rCBV values in PWI are suggestive of HGG with the exception of tumors with a oligodendrocytic component that present with high rCBV values regardless of tumor grade [2].

Primary CNS lymphoma usually presents as a homogenously enhancing, well circumscribed, lobar or periventricular lesion. Multiple lesions exist in up to 30 % of primary CNS lymphoma patients. Elevated choline, lipid and lactate signals, and reduced NAA signal in MRS are typical for brain lymphoma and can help differentiate it from toxoplasmosis in AIDS patients. Lymphomas also tend to have lower ADC and rTBV values in DWI and PMRI compared to gliomas [1, 3].

Tumor mimicking lesions such as tumeffactive demyelinating lesions (TDL) and brain abscesses can have imaging characteristics that are very similar to HGGs or metastases. Brain abscesses have a distinctive pattern of signals in MRS and TDL rarely shows highly elevated rCBV in PWI.

2 Monitoring of the Therapeutic Response of Brain Tumors

MRI is again the imaging modality of choice. Serial imaging controls should be performed in all cases and types of brain malignancy. Cerebral metastases require an imaging follow-up every 3 months for at least 2 years after treatment and the same protocol applies in most HGGs and CNS lymphomas. Follow-up periods of 6 months can be considered if no tumor recurrence has occurred within the first 2 years. LGGs can undergo imaging controls every 6 months for 2 years after an initial postoperative MRI. Yearly controls can be considered after 2 years without tumor recurrence.

Imaging assessment of a brain tumor after surgery or irradiation has always been a challenge for neuroradiologists, neurooncologists, and neurosurgeons. Postoperative changes in the tumor bed and radionecrosis can be mistaken for tumor remnants or recurrence. DWI can be a very useful instrument in differentiating postoperative cellular damage and subsequent contrast enhancement from tumor recurrence, especially in patients receiving adjuvant radio- and chemotherapy [2]. MRS and DWI can also be of some value in recognizing radionecrosis, which exhibits a lower choline signal and higher mean ADC values compared to pure tumor. Finally, PET and SPECT scans have a rather well established efficacy in distinguishing radionecrosis from tumor recurrence.

An important step forward in the accurate assessment of response to therapy in patients with malignant gliomas is the recent introduction of criteria of the response assessment in neuro-oncology (RANO), an improvement of the 1990 established Macdonald criteria [4]. They are based on the product of the maximal cross-sectional diameter of an enhancing lesion but the non-enhancing tumor component is also considered. The RANO criteria take into account new treatment options and changes in imaging procedures. In contrast to the previous criteria measurable (well demarcated) and non-measurable lesions (blurred or cystic) are defined, which are especially important to determine endpoints in clinical studies. If the response rate is the primary endpoint of the study patients with measurable disease are required. If the duration of tumor control is the endpoint of the study, then patients with measurable and non-measurable lesions are eligible. Furthermore, for the first time, tumor size is additionally measured on T2-/FLAIR-weighted images and progress is defined by a significant increase of the T2-/FLAIR lesion.

3 Conclusion

The imaging modalities of the central nervous system available to a physician today are many and can provide a considerable amount of information about its anatomy, physiology, and function. Recent advances in computational power and software sophistication have made this information readily available for treatment planning, guidance, and follow-up. In order to improve the quality and diversity of the anatomic imaging, efforts to develop MRI systems with field strength higher than 1,5 Tesla have been made since the late 1980s. High field (3 Tesla) MRI has been available in clinical practice since 2002 and a number of these systems, with a higher spatial resolution and speed, are now operational in many hospitals around the world and tend to become the new standard. These systems provide images of unsurpassed anatomical detail, a feature extremely useful in tumor detection and surgical treatment. Ultra high field (up to 12 Tesla) MRI is currently developed or evaluated in certain neuroscience research facilities mainly in Europe and North America. Technical improvements in electromagnetic source imaging (ESI), i.e., the combination of magnetoencephalography (MEG) and electroencephalography (EEG)-have also made this real-time functional brain mapping technique a valuable complement to the fMRI.