The utility of thyroglobulin

This is an example of the utility of thyroglobulin. This is a patient who had surgery and 131 iodine ablation over here. The thyroglobulin level was initially around 5 at the time they were off suppression. They were started on thyroxin suppression. Their TSH was brought low and their thyroglobulin declined. A year later their thyroxin was stopped, TSH went back up, there was a slight increase in the thyroglobulin levels but the 131 iodine scan was negative, so they were put back on suppression. Several months after that their thyroglobulin, even while on suppression, started to rise and then when they were taken off suppression there was a considerably greater increment in their thyroglobulin, and the scan at that time was positive showing recurrent disease with pulmonary mets, which was treated with a second treatment with radioiodine.

Finally, as far as the other treatment issue that is available is the question of what one should do as far as thyroxine suppression. This is just some data from a recent, retrospective study, in which patients were scaled regarding the degree of thyroxine suppression. There were some patients who had essentially undetectable TSH’s and other patients whose TSH was not quite fully suppressed. What this is showing is that in patients with – this is going back to a TNM staging – TNM stage I or II disease, that there was not very much of an impact on recurrence related to the degree of thyroxin suppression but that in these higher risk patients with stage III disease, particularly, that the patients who were more fully suppressed had a lower risk of recurrence. So again, certainly in some of the patients who might be at a higher a priori risk there’s justification in treating them with sufficient thyroxine to keep their TSH suppressed, I feel that to an undetectable level at least for the first several years after treatment.

What about the patients who, in a way, I suppose you might be more likely to see, or the ones we need to call for additional help on, are those with differentiated radioiodine resistant thyroid cancer. A couple of potential treatment options; there is a role for external beam irradiation in these patients.

Chemotherapy has been, I’d say, disappointing at best, but there still may be a role for it. The most commonly used drug has been Adriamycin, either by itself or in combination with cisplatin, with other agents. There are a variety of kind of interesting approaches being tried right now. In addition I believe there is a Taxol trial going on, although I haven’t seen any results on it. There has also been this interest in redifferentiation, in trying to treat the tumor in such a way as to lessen its radioiodine resistance and enable it to be more effective in taking up radioiodine. Retinoic acid and its congeners has been used and there has also been this interesting finding that occasionally in patients treated with Adriamycin as chemotherapy, that if you repeat the scan after they have gone through a couple of treatment cycles, that there have been some patients who have then been shown to take up radioiodine.

Suggesting that in some way there was some type of redifferentiation that took place. But clearly, these patients who get to a stage of radioresistant disease are a very very troubling group to treat.

My approach, in terms of these patients, is; the question is whether they have progressive disease and whether they are symptomatic. If they are not symptomatic – and many many patients, even with extensive thyroid cancer, may not be – or maybe minimally symptomatic, there is certainly reason to simply continue to follow those patients. If there are significant symptoms taking place, which may frequently relate to bony metastases, then I think the important issue is to localize the disease. If the disease if very focal, if it involves an isolated or bony matter or a large soft tissue lesion, then there is certainly a role for external beam irradiation and/or surgery, depending on the locations involved. If the patient is progressing, has diffuse disease, then I think that might be the situation in which there would be a role for systemic chemotherapy, using either Adriamycin or some combination of agents. Then as I mentioned, consider at least not giving up totally on radioiodine but consider another scan maybe after a few cycles to see if there might have been some redifferentiation issue.

MRI initially provided two types of images

MRI initially provided two types of images, designated T1 and T2. For brain tumors, the former generally showed a well-demarcated area of low density, and the latter showed bright whiteness that encompassed a more extensive region owing to the signal of the surrounding brain edema. With the availability for general usage in 1988 of gadolinium contrast for MRI, a new set of criteria of usage and differential diagnostic considerations in brain imaging have quickly evolved. T1 gadolinium imaging is the most precise way to image a brain tumor, and patients can often be followed up during and after treatment with that type of study alone. Such an approach is easier for patients because it reduces the length of time otherwise spent on T2 scanning. Now and then, T2 images are useful. For example, T2 images, besides showing the extent of edema, also delineate the demyelinating effects of radiation on white matter. FLAIR images, a variant of T1, are even better for this.


Metastases These are remarkably variable.
Some enhance brightly and solidly with gadolonium. Others are in ring
configuration. Many are invisible with contrast CT.
Acoustic neuromas These are invariably intensely
contrasted by gadolinium, even more reliably than by CT.
Meningiomas Same as for acoustic neuromas.
Pituitary adenomas These always enhance less than
the normal pituitary gland. MRI is superior in every way to CT,
especially when thin slices and magnified views are ordered.
Glioblastoma These are almost always in ring
Anaplastic astrocytomas These are sometimes solidly
bright; they are often patchy, may be noncontrasting, and may look like
low-grade astrocytoma.
Low-grade astrocytomas These do not enhance. They are
often invisible by CT or are imaged only as vague low density.
Oligodendrogliomas These generally do not enhance
unless anaplastic and are often invisible on CT unless they are
Primary brain lymphomas These usually exhibit
homogeneous enhancement and are smoothly rounded. Periventricular
location is common. They are multiple in about a fourth of cases. This
lesion does not often look like glioblastoma but is easily mistaken for
metastases if multiple.

Cerebral angiography seldom is used in the diagnosis of brain tumors. In a few circumstances, neurosurgeons, in preparation for surgery, require a more precise knowledge of the pattern and position of blood vessels, which can be obtained only by angiography. The procedure is also used to embolize highly vascular meningiomas or to study cerebral dominance by injection of barbiturate into the carotid artery (the Wada test) in left-handed individuals who are to have surgery near language areas. Preoperative determination of cerebral localization helps surgeons to plan the extent of surgery and to avoid creation of postoperative language deficits in the patient.

Examination of the spinal fluid has limited indication in the diagnosis of brain tumors. One is to rule out an inflammatory disorder mimicking a brain tumor. Another is to establish the diagnosis of benign intracranial hypertension in patients with uninformative MRIs. In addition, spinal fluid cytology may be useful for determining instances of malignant meningitis secondary to metastatic neoplasms in association with spinal spread of medulloblastoma in some children and in identifying primary lymphomas of the brain in cases in which MRI changes are ambiguous.

The routine electroencephalogram (EEG) has no role in the diagnosis of brain tumors and does not assist in the choice of anticonvulsant drugs for patients with brain tumor. However, specialized intraoperative neurophysiologic techniques, such as depth electrode studies and intraoperative monitoring, may be useful in identifying and removing epileptogenic areas adjacent to brain tumors or to avoid resection of critical brain regions adjacent to tumors.

Positron emission tomography (PET) is able to quantify biochemical functions, such as oxygen and glucose utilization, within tumors as well as in normal brain tissue. PET scanning is a powerful research tool of limited availability for routine clinical purposes. Its spatial resolution is inferior to that of both CT and MR. In patients with brain tumors who develop recurrent symptoms after radiation therapy, PET can differentiate, with about 70% accuracy, radiation-induced injury from tumor recurrences. These disorders often appear identical on MRI.

Differential diagnosis of brain tumors

Patients who present with symptoms and signs of increased intracranial pressure or a first convulsive seizure need to be hospitalized. Diagnosis and treatment measures must be started at once; it may be unsafe to wait. Those who present with focal neurologic impairment and who do not have symptoms of increased intracranial pressure may reasonably be evaluated in the outpatient setting for other conditions that are often considerations in the differential diagnosis of brain tumor. The tempo of evolution of symptoms and signs of focal neurologic impairment, much more than their severity, governs urgency of evaluation. The tempo also strongly influences diagnostic considerations. Although an occasional brain tumor may manifest with such rapid onset of hemiparesis or aphasia that a stroke is mimicked, most do not. Associated aspects of the history, such as recent head trauma, previous episodes of reversible neurologic impairment, or recent infection and fever, should direct attention to diagnostic alternatives such as subdural hematoma, multiple sclerosis, or cerebral abscess. Simply stated, it is the careful history, not the neurologic examination, that usually points to the alternative diagnoses.


Brain imaging by MRI or CT scans is an indispensable component of the modern diagnosis of the presence, but not the type, of brain tumors. One type of tumor can look like another or even resemble a non-neoplastic mass lesion, such as a brain abscess, fungal infection, parasitic invasion, demyelinating disease, or stroke. For definitive diagnosis and adequate treatment planning, one must obtain a tissue diagnosis whenever possible. This can be made either by direct surgical biopsy or, in the case of some non-neoplastic conditions, by judging CT or MRI responses to particular therapies.

MRI is almost always superior to CT scanning in diagnosing intracranial mass lesions. MRI outlines posterior fossa structures and tumors with a clarity that CT cannot achieve because of x-ray distortions caused by the bony structure of that region. In several types of tumor, particularly the low-grade gliomas, MRI may show extensive brain infiltration in cases that fail to produce any image abnormality on CT or, at most, show a vague area of low density. Although either MRI or CT should be used with contrast enhancement in cases of suspected brain tumor, the passage of such contrast agents beyond the blood-brain barrier into the tissue does not necessarily imply the presence of a histologically malignant tumor. For example, although malignant gliomas almost always show contrast enhancement, so do meningiomas, which are entirely benign if they can be fully removed surgically.

CT scans done without contrast enhancement are of little value in the diagnosis of brain tumors or other mass lesions. Although it is true that hemorrhage, calcifications, hydrocephalus, and shift can be well seen on a non-contrast CT scan, the interpretation of even these conditions is tentative because each can have an underlying causative structural abnormality, such as a brain tumor, which may fail to appear on a non-contrast CT study. Allergy to CT dye is rare and is readily manageable. Currently available non-ionic CT dyes have an extremely low incidence of side effects. Currently used CT dyes carry little risk of causing renal dysfunction in normally hydrated patients who are not known to have kidney disease.


Hematomas, especially in tumors
that have a tendency to bleed, such as melanoma
Abscesses, including fungal
Parasitic infections, such as
Vascular malformations,
especially those without arteriovenous shunts
Solitary large plaques of
multiple sclerosis
Progressive strokes (rare)

Initial evaluation


Brain tumors present in two patterns, not necessarily mutually exclusive. One consists of nonfocal symptoms of increased intracranial pressure, such as headaches, nausea, vomiting, confusion, and lethargy. The other consists of symptoms or signs of focal brain dysfunction, such as hemianopia, hemiparesis, cranial nerve palsies, or focal seizures. Such signs of focal brain dysfunction may have convincing localizing value even before an image of the brain is made by computed tomography (CT) or magnetic resonance imaging (MRI).

Some tumors that arise in neurologically “silent” areas, such as the parietal or frontal association cortices, may produce only nonfocal generalized symptoms of headache, confusion, behavioral change, or, eventually, a seizure, despite growing to a considerable size. Although the capacity to reach early diagnosis by CT or MRI has greatly reduced the numbers of patients in whom symptoms of increased intracranial pressure represent initial complaints, examples still remain, especially in association with fast-growing tumors and in children. The latter are particularly likely to have tumors in the posterior fossa that tend to obstruct spinal fluid pathways earlier than do supratentorial tumors. The tempo with which a brain tumor grows also influences the presenting symptoms. Despite the fixed space within the skull (once infantile sutures have closed), the human brain possesses a remarkable capacity to make room for a slowly growing tumor. Because of this, and even allowing for the relative rapidity of growth of aggressive brain tumors, such as glioblastomas, the patient usually appears better clinically than might be expected from the degree of abnormality seen on CT or MRI scan.


  • Frontal lobe
  • Generalized
  • Focal motor
    seizures (contralateral)
  • Expressive aphasia
    (dominant side)
  • Behavioral changes
  • Dementia
  • Gait disorders,
  • Basal ganglia
  • Hemiparesis (contralateral)
  • Movement disorders
  • Parietal lobe
  • Receptive aphasia
    (dominant side)
  • Spatial
    disorientation (nondominant side)
  • Cortical sensory
    dysfunction (contralateral)
  • Hemianopia (contralateral)
  • Occipital lobe
  • Hemianopia (contralateral)
  • Visual disturbances
  • Temporal lobe
  • Complex partial
    (psychomotor) seizures
  • Generalized
  • Behavioral changes
  • Olfactory and
    complex visual auras
  • Corpus callosum
  • Dementia (anterior)
  • Behavioral changes
  • Asymptomatic (mid)
  • Thalamus
  • Sensory loss (contralateral)
  • Behavioral changes
  • Language disorder
    (dominant side)
  • Midbrain/pineal
  • Paresis of vertical
    eye movements
  • Pupillary
  • Precocious puberty
  • Sella/optic nerve/pituitary
  • Endocrinopathy
  • Bitemporal
  • Monocular visual
  • Pons/medulla
  • Cranial nerve
  • Ataxia, nystagmus
  • Weakness, sensory
  • Spasticity
  • Cerebellopontine angle
  • Deafness (ipsilateral)
  • Loss of facial
    sensation (ipsilateral)
  • Facial weakness (ipsilateral)
  • Ataxia
  • Cerebellum
  • Ataxis (ipsilateral)
  • Nystagmus

Common brain tumors


Lung (37) Meningioma (80) Glioblastoma (47)
Breast (19) Acoustic neuroma (10) Anaplastic astrocytoma (24)
Melanoma (16) Pituitary adenoma (7) Astrocytoma (15)
Colorectum (9) Other (3) Oligodendroglioma (5)
Kidney (8) Lymphoma (2)
Other (11) Other (7)

These figures, given in parentheses, can be extremely variable from one center to another, depending on referral pattern. They are given here as general estimates based upon many published series.


A. Astrocytic tumors
1. Astrocytoma
2. Pilocytic
3. Subependymal
giant cell astrocytoma (ventricular tumor or tuberous sclerosis)
4. Astroblastoma
5. Anaplastic
(malignant) astrocytoma
B. Oligodendroglial tumors
2. Mixed
3. Anaplastic
(malignant) oligodendroglioma
C. Ependymal and choroid plexus
a.Myxopapillary ependymoma
b.Papillary ependymoma
2. Anaplastic
(malignant) ependymoma
3. Choroid plexus
4. Anaplastic
(malignant) choroid plexus papilloma
D. Pineal cell tumor
1. Pineocytoma (pinealcytoma)
2. Pineoblastoma (pinealoblastoma)
E. Neuronal tumors
1. Gangliocytoma
2. Ganglioglioma
4. Anaplastic
(malignant) gangliocytoma and ganglioglioma
5. Neuroblastoma
F. Poorly differentiated and
embryonal tumors
1. Glioblastoma
a.Glioblastoma with sarcomatous component (mixed glioblastoma and sarcoma)
b.Giant cell glioblastoma
2. Medulloblastoma
a.Desmoplastic medulloblastoma
4. Primitive polar
5. Gliomatosis

This is one of several formal schemes that are based on neuropathologic criteria. Metastasis is not considered, and one can get no sense of a given tumor as a clinical problem, as suggested by the simple classification in previous table.

Brain Tumors

More than 18,000 new cases of primary brain tumors are treated each year in the United States. Metastases are even more frequent and contribute considerably to suffering and death from systemic cancer. The diversity of brain tumors makes it important to attend to what is characteristic about each histologic type. Biologic specificity guides therapy to some extent now, and will be the key to successful treatment in the future.

The classification of brain tumors is a subject with confusing terminology. This text employs the simple approach of classifying brain tumors into metastatic, primary extra-axial, and primary intra-axial . These categories include all of the primary brain tumors listed in the World Health Organization classification , and adds pituitary and metastatic tumors. Although obviously simple, it follows practical clinical thinking. This chapter deals with the general biology, clinical features, and treatment of brain tumors as an overall problem.


“Is it benign or malignant?” is invariably the first question asked by patients, families, and physicians when confronted with a diagnosis of brain tumor. About a third of primary brain tumors can be called benign. Meningiomas and acoustic neuromas are good examples. They grow slowly, often can be removed completely, and rarely recur.

The concept of malignancy in the central nervous system (CNS) has a different meaning from that which applies to systemic cancers. The term “malignant” has nothing to do with metastasis out of the CNS, which is extraordinarily rare. It has everything to do with anatomic location and the possibility of complete surgical removal. Unless a tumor can be completely excised to the last cell, all intracranial neoplasms are potentially malignant in that they may recur, and often do.

Osmotic Agents

Additional therapy for increased ICP includes the use of osmotic diuretics, such as mannitol. In the face of deepening coma, pupil inequality, or other deterioration of the neurologic examination, mannitol may be life saving. Mannitol (0.25 to 1.0 gm/kg) can effectively reduce cerebral edema by producing an osmotic gradient that prevents the movement of water from the vascular space into the cells during membrane pump failure and draws tissue water into the vascular space. In effect, this reduces brain volume and provides increased space for an expanding hematoma or brain swelling. The osmotic effects of mannitol occur within minutes of its administration and peak at about 60 minutes after the bolus has been administered.

The ICP-lowering effects of a single bolus may last for 6 to 8 hours. Mannitol has many other neuroprotective properties. It is an effective volume expander in the presence of hypovolemic hypotension and therefore may maintain systemic blood pressure required for adequate cerebral perfusion. It also promotes CBF by reducing blood viscosity and microcirculatory resistance. Mannitol reduces RBC deformity and therefore improves oxygen carrying capacity. It is an effective free radical scavenger, reducing the concentration of oxygen free radicals that may promote cell membrane lipid peroxidation.

Pancreatic cancer information