Glioblastoma is a very complicated and heterogeneous disease. Even individual tumors may have tumor cells with very different characteristics, most notably the signaling cascades. These signaling cascades are an important component of normal cell function and are important in maintaining the integrity of our varied organs. However, in the case of cancer such as glioblastoma, many of these signaling cascades are functioning abnormally and do not maintain the normal controls. For example, p53 is involved in the signaling pathway that is responsible for programmed cell death, an important normal function that allows old cells to be replaced by new cells. In many gliomas, the p53 gene has a mutation that reduces or loses its function so that these tumor cells do not have the mechanism to respond to injury by undergoing programmed death (apoptosis). This is only one example of a mutation that results in abnormal pathway function. There are many other mutations causing abnormal signaling cascade function such as isocitrate dehydrogenase 1 (IDH1) involved in energy production, epidermal growth factor receptor (EGFR) resulting in abnormal tumor cell proliferation and resistance to death, and a variety of mutations in other cascades leading to abnormal function. Additionally, some cascades function abnormally as a result of over expression of genes, rather than a mutation of a gene. Examples include EGFR as well as the platelet derived growth factor receptor (PDGFR) and others. Each individual glioblastoma tumor can be characterized by having multiple abnormalities in the signaling cascades, thereby making it challenging to find the most important ones to target with treatment.

Tumor measurements during active treatment, typically using magnetic resonance imaging (MRI) are a critical component of successful patient care. On rare occasions when patients are unable to undergo MRI, computed tomography (CT) is used. Brain tumors are very complex and this is accentuated by changes on MRI that are caused by treatment. For example, recent studies have shown that soon after completion of concurrent radiation and temozolomide for glioblastoma, nearly half of the patients have imaging studies that show an increase in the size of the abnormality. However, many of these patients do not have tumor progression, rather changes that are directly caused by the treatment. These changes, called “pseudoprogression” are indistinguishable from tumor progression. Making the distinction is typically done by continuing treatment and providing frequent tumor imaging. Pseudoprogression will improve over time, whereas true tumor progression will continue to worsen.

To date, there are no blood markers that have been shown to be useful in monitoring tumor activity as exists for other cancers such as the PSA for prostate cancer.

Typically we perform what is referred to as phase I testing to determine the optimal chemotherapy dose and dosing schedule. For most chemotherapy, the highest tolerable dose has proven to be the most effective and most of the phase I studies do strive to determine the dose that is both tolerated and most effective. However, some of the newer “targeted” therapies may have the desired effect at lower doses so there are now more studies that are looking at tumor tissue after treatment to see if the effective dose can be lowered, thereby reducing the risk of side effects.

The treatment of recurrent glioblastoma has been challenging. Recently, bevacizumab (Avastin) was approved as a treatment for patients with recurrent glioblastoma. Although many patients do benefit from this treatment, tumor regrowth is common so there needs to be additional effective therapies for recurrent disease. This is a major area of research in many centers, consortia, and cooperative groups. One of the most active areas of research is focused on studying the changes in tumors from the time of diagnosis to the time of recurrence. This is important to ensure that targeted treatments will be directed at the targets present in the current tumor and not rely on information that may no longer be accurate. This avenue of investigation will require a concerted and collaborative effort that is now underway in several groups.

Recent studies over the past 5-7 years has underscored that there are several distinct subtypes of glioblastoma. However, the definitions of these subtypes vary by the type of evaluation and the criteria used to make the subtype classification. For example, there is clear evidence that glioblastoma may arise directly as a grade IV tumor (primary glioblastoma) or evolve from a lower grade (II or III) glioma (secondary glioblastoma). These 2 subtypes have distinctive molecular profiles, suggesting that optimal treatment may vary by this distinction. However, there are additional studies suggesting that there may be ways to categorize glioblastoma based on features from tumor profiling. The recent results from the Tumor Genome Atlas (TCGA) demonstrated that there were 4 subgroups: mesenchymal, proneural, neural and classical. Although distinct molecular profiles categorize these groups, these designations have not yet been successfully used to define subgroup specific treatment. Other molecular markers such as MGMT gene promoter methylation status may help predict prognosis. However, to date, with the exception of the chromosome changes with deletion of 1p and 19q in anaplastic oligodendroglioma, there are no markers that help in deciding optimal treatment (predictive marker).

After completion of the initial treatment regimen, all patients who have been successfully treated for a malignant brain tumor require routine surveillance. In my practice, I evaluate patients with high grade (III or IV) tumors every 2 months for a year, then every 3 months for a year, every 4 months for a year and then every 6 months from that point on. Of course, any changes in clinical status warrants an urgent evaluation even if it is earlier than the evaluation plan.

Targeted therapy can refer to a wide range of treatments. It is often used in the context of cancer treatment to refer to a therapy that targets a marker or characteristic of the cancer cell that is either unique or at least different enough from normal cells to provide a therapeutic advantage. There are very dramatic examples in other cancers such as chronic myeloid leukemia where a characteristic chromosome change (referred to as the Philadelphia chromosome) results in a unique protein found only in tumor cells. Targeting this protein (imatinib, Gleevec) results in dramatic cancer responses. Unfortunately, there are few “unique” targets in malignant brain tumors although there are many examples of overexpression (high levels) of certain proteins and activation of molecular pathways that are potential treatment targets. To date, likely because of multiple abnormal molecular pathways, targeting a single component of a pathway has not resulted in significant impact of tumor growth. More recent clinical trials are evaluating combinations of these “targeted” drugs and carefully studying the effects of treatment on the tumor cells providing an opportunity to uncover the reasons for treatment success or failure.

The treatment for brain tumors does depend upon the tumor type. This emphasizes the need for an accurate pathologic diagnosis. With the exception of some tumors in the brainstem, all tumors should undergo either a tumor resection or biopsy so the pathologist can determine the type of tumor. For glioblastoma, also known as a grade IV astrocytoma, the oral chemotherapy agent temozolomide is used in combination with radiation treatment. After the radiation is completed, then patients typically receive 6 to 12 months of additional temozolomide chemotherapy. The role of chemotherapy is less clear for other brain tumors. However, the late results of 2 studies did determine that radiation and chemotherapy is optimal treatment for patients with anaplastic oligodendroglioma tumors that contain the 1p 19q chromosome loss. The role of chemotherapy for low grade (grade II) gliomas and anaplastic astrocytomas (grade III) has not yet been clearly defined.