Approximately 80 million CT scans are performed in the United States every year. Adrenal tumors are found incidentally in about 5% of these scans. Most of these tumors will turn out to be benign, but a small fraction will be adrenal cortical carcinoma (ACC), a type of cancer with high mortality and frequent recurrence; even for localized disease the 5-year survival rates do not exceed 65%, and distant spread is associated with a >90% death rate within that time period. Early diagnosis of a malignant adrenal mass is therefore imperative to assure timely and appropriate therapy.
Unfortunately, CT imaging alone is very limited in its ability to distinguish benign from malignant adrenal tumors; only very small and hypodense lesions can be easily dismissed as benign. The sizeable group of patients with larger or denser tumors end up with an arduous workup that frequently includes additional imaging studies, hormonal testing, and biopsy. However, even the latter has both a high diagnostic false positive and false negative rate, and ultimately the tumor is often resected, often unnecessarily, while, on the other hand, the delays due to the diagnostic work might also compromise optimal care for those tumors that prove malignant.
In addition, patients that are believed to probably not have adrenal cancer after their workup, and those who opt out of biopsy or surgery, often still require long-term follow up with regular re-imaging and repeated hormone testing, with resultant radiation exposure and high health care costs.
Mayo Clinic’s Clinical Mass Spectrometry Laboratory (CMSL) has added a new, noninvasive, and more accurate test to diagnose malignant adrenal tumors, via urinary steroid profiling. The test is the result of several years of analytical and clinical studies, some of which have been published in the journal Clinical Chemistry, and is based on cutting edge liquid chromatography, high-resolution, accurate-mass mass spectrometry (HRAM LC-MS) measurement for 25 steroid metabolites in urine.
“Our new test for adrenal cortical carcinoma will differentiate this rare and lethal tumor from benign adrenocortical adenomas (ACAs), including those that overproduce corticosteroids, or mineral steroids, or sex steroids, or those that are hormonally inactive. We also anticipate that the test will be able to aid in the diagnosis of inborn errors of steroid metabolism, such as congenital adrenal hyperplasia (CAH),” says Ravinder Singh, Ph.D., Co-Director of CMSL and senior author of the paper.
“The test will utilize both clinical and laboratory data. The clinical parameters are age at diagnosis and sex of the patient, the size of the tumor by CT scanning and its density in Hounsfield units, whether it was detected incidentally or not, and whether there is evidence of hormone overproduction. All these data are readily available for almost all patients with an adrenal mass, and are used by our algorithm to calculate the pretest probability of having ACC. The steroid profile testing is then performed and its results are added into the risk calculation algorithm to generate a posttest probability. The final result will provide the referring physicians a highly accurate probability for ACC and will thereby facilitate the optimal choice of further investigation, if any, based on an informed discussion between doctor and patient.”
“In clinical diagnostics, we want to minimize the risks of false-positive or false-negative results,” says Dr. Singh. “Thanks to HRAM LC-MS technology, resolution is now high enough to allow us to measure many steroids simultaneously, instead of testing only a single steroid, and progress in complex data analysis has allowed us to combine all this data with clinical parameters to arrive at accurate ACC risk estimates.”
The human body has two adrenal glands, one above each kidney. Adrenal glands influence many processes and functions of our body, mainly through production of three types of steroid hormones:
These steroids are all synthesized from cholesterol, via enzymes in the adrenal glands. In benign ACAs, near-normal levels of precursor and bioactive steroids are produced. By contrast, ACC frequently shows abnormal patterns of steroid production. By measuring 25 different steroid metabolites, even subtle abnormalities can be detected. This is the basis for the diagnostic capability of 25 steroid profiling .
The timing for this new biomarker comes none too soon, since, as aforementioned, adrenal masses are found in 5% of the population. The prevalence increases with age, ranging between less than 0.5% in children and around 10% in 70-year-old patients.
“Radiologists are finding adrenal nodules incidentally all the time, during patient examinations for reasons other than evaluation for adrenal disease,” says Irina Bancos, M.D., an endocrinologist in Mayo’s Department of Internal Medicine, Division of Endocrinology, who co-authored the paper. “Although the majority of these tumors are benign, around 30% of adrenal tumors (>4cm) are malignant (most represented by ACCs), and the survival rate for these patients is very poor unless detected early.”
To conduct urinary steroid profiling, the investigators have used high resolution mass spectrometry. This allows the simultaneous detection of large numbers of different steroids, despite the fact that many of these compounds have the same, or very similar, molecular structures and molecular weights. At the same time, nonspecific interferences are also dramatically reduced and signal to noise ratios are improved, resulting in high analytical sensitivity and specificity.
“It allows us to distinguish between steroids with at least twice the accuracy of low-resolution mass spec,” says Dr. Singh. “In clinical diagnostics, especially ACC, high accuracy, sensitivity and specificity are very important. We don’t want to put any unnecessary risks onto our patients.”
Although the use of high resolution is more costly than low resolution, the greater margin for error on a low-resolution instrument “can turn out to be quite expensive for a cancer patient,” adds Dr. Singh.
As it stands, clinicians must rely on their acumen to determine the likelihood of an adrenal tumor being malignant or benign. This is a complex decision, which depends on tumor size, imaging characteristics, and production of a few steroid hormones that can be routinely tested (typically, existing lab tests do not add additional diagnostic values for confirming ACC).
Larger malignant adrenal tumors are more readily recognized as malignant based on size, heterogeneity, areas of necrosis, and other imaging characteristics. It’s the smaller adrenal tumors that can present a diagnostic challenge, because when examined with current imaging technologies (e.g., CT scans and MRIs), benign adrenal tumors often “mimic” malignant ones.
“Clinical assessment of probability for malignancy works relatively well when expert physicians are involved and the tumors are relatively large,” says Dr. Bancos. “Having evidence-based clinical algorithms can in part improve this situation by standardizing risk assessment and enabling less specialized physicians to make better diagnostic decisions. However, for small and medium-sized tumors, even the experts are still frequently struggling to distinguish ACC from ACA or from other cancerous lesions in adrenal glands (such as sarcomas, metastases, lymphomas, etc). This is why repeat imaging and extensive follow-up, and in some cases, adrenal biopsy or surgical exploration are often needed to arrive at a definitive diagnosis.”
Because of this uncertainty, a significant number of adrenal biopsies and adrenalectomies are performed on suspicious looking adrenal tumors unnecessarily, ultimately showing a benign tumor. On the other hand, adrenal biopsy of ACC is not advised because of several reasons.
“Adrenal carcinoma has a capsule holding it together and, theoretically, this prevents the cancer from spreading locally,” says Dr. Bancos. “By biopsying it, we would open a door to dissemination, and that may significantly deteriorate the prognosis. Moreover, adrenal biopsy has a suboptimal accuracy for ACC and is usually reserved for other cancers. Combining algorithmic clinical risk assessment with steroid profiling allows us to diagnosis ACC in a noninvasive manner.”
Any time a cancerous lesion is found in an adrenal gland, clinicians have to differentiate between two scenarios:
“In the first situation, steroid profiling would allow us to diagnose adrenal cancer in a noninvasive manner, in which case a biopsy would be avoided and surgery would be scheduled ASAP,” says Dr. Bancos. “In the second situation, where cancer has travelled to the adrenal gland from somewhere else, steroid profiling will not tell us what kind of cancer it is. For example, is it lung cancer metastasis to the adrenal gland in a smoker, or colon cancer metastasis to the adrenal gland in someone without a colonoscopy? Or is it melanoma metastasis to the adrenal gland? In this situation, biopsy may be valuable, because it would tell us what type of primary cancer we are dealing with.”
With a simple, noninvasive urine test combined with clinical data that are regularly available, the new biomarker panel allows clinicians to make the diagnosis right away on these smaller, indeterminate tumors, and thus avoid unnecessary follow-up imaging visits, unneeded biopsies, or even adrenalectomy, where the entire mass is removed to reach a diagnosis.
“It goes without saying that a negative biomarker result should also avoid substantial health care costs,” says Dr. Bancos. “Conversely, in patients with a small ACC, a positive biomarker result will lead to an earlier intervention and will radically improve patient prognosis, because we would be removing a smaller cancer before it has a chance to grow and metastasize.”
The assay also promises to alleviate patient anxiety—patients and their families won’t have to wait so long, over repeated health care provider visits and tests, for an accurate diagnosis. It would help diagnose other adrenal diseases as well.
“We have promising preliminary data that this steroid profile could also be useful for the diagnosis and differential diagnosis of a number of functional adrenal diseases, from CAH to various syndromes of overproduction of certain bioactive steroids, such as cortisol or aldosterone,” says Stefan Grebe, M.D., Ph.D., CMSL/DLMP, who also co-authored the paper.
It would be overly expensive and complicated for most other institutions to implement this kind of metabolomics testing. Fortunately, the new assay will soon be available through Mayo Clinic Laboratories (MCL), the commercial laboratory testing wing of Mayo Clinic. Relevant clinical data will be collected at order entry, and the steroid profile data will be combined with the clinical risk data to arrive at an accurate risk ACC risk assessment.
“MCL has historically been a leader in performing a wide range of endocrine testing,” says Dr. Singh. “We’re experts at using liquid chromatography and mass spectrometry, and now, we’re moving into this novel testing profile, which will be one of the first metabolomics multiple-analyte biomarker panels for cancer and will also pioneer algorithmic fusion of clinical and laboratory data in the cancer field.”