Prostate MRI adds brains to cancer treatment brawn

Armed with endorectal coil MRI and other advanced imaging techniques, radiologists could one day help transform management of the most common cancer in men, saving many from needless impotence, incontinence, and post-treatment recurrence.

One in six men will get prostate cancer in his lifetime, and incidence is expected to increase with the graying of the baby boomers. It is anticipated that about 219,000 men will be hit with a diagnosis this year.

The indolent disease accounts for 9% of estimated cancer deaths in men, the second-highest number behind lung and bronchus deaths (31%).

Imaging techniques could help improve prostate cancer management, which today has many drawbacks. In particular, MR imaging with an endorectal coil could have a dramatic impact on detection, localization, treatment planning, and monitoring.

"I think the 1.5T [prostate] exam is close to being a very robust routine clinical exam today, and it will make a real difference over the next couple of years," said John Kurhanewicz, Ph.D., director of the prostate imaging group and the biomedical and MR laboratory at the University of California, San Francisco.

Prompted by the Boston-based advocacy foundation AdMeTech, a bipartisan House bill is now drawing Congress's attention to the role of radiology in prostate care and to calls for greater research funding.

"Lack of accurate imaging tools means biopsies can miss cancer even when multiple samples are taken, and current treatments-either radical surgery or radiation-can leave 50% to 80% of men incontinent, impotent, or both," the bill states. "With imaging tools, millions could be saved in improved detection and treatment."

AVOIDING IMPOTENCE

Today's screening and biopsy methods have limitations, but they are still much better than in the past. Cancers are found at ever earlier stages, metastatic disease is seen much more rarely, and more men are eligible for minimally invasive treatments.

Thanks to advances in early detection, men now may have 20 to 30 years to live with the outcomes of various therapeutic choices, according to researchers at Memorial Sloan-Kettering Cancer Center in a recent overview of state-of-the-art prostate cancer imaging (Radiology 2007:243[1]:28-53).

After a positive biopsy, urologists now want to know if they can opt for active surveillance instead of immediate treatment, Kurhanewicz said. He noted that 1.5T MR can really have an impact on this important question.

At Memorial Sloan-Kettering, MRI is used routinely to monitor patients on active surveillance, said Dr. Hedvic Hricak, head of the radiology department. Furthermore, MRI is frequently used before men go under the knife.

"MRI provides a road map. Why go in blindly if you can predict exactly where a tumor is and plan surgeries with MRI?" she said.

NEW TREATMENTS

Treatment options are now plentiful. Robotically assisted laparoscopic radical prostatectomy (RALRP), for example, is starting to replace radical prostatectomy. Ideally, RALRP is performed on patients with a life expectancy of greater than 10 years and a diagnosis of organ-confined cancer or limited extracapsular disease.

Alternatives to surgery include high-intensity focused ultrasound, brachytherapy, and cryotherapy.

To guide treatment decisions, it is important to assess whether the disease is confined to the organ or has spread beyond the capsule of the prostate gland (extracapsular extension) to the seminal vesicles or lymph nodes.

Traditionally, bone and CT scans have been performed on the newly diagnosed, but these studies are insensitive and not suitable for assessing men with early-stage disease or for predicting treatment success, said Dr. Stephen Strum, an oncologist and cofounder of the Prostate Cancer Research Institute.

Rising prostate-specific antigen after treatment, called PSA recurrence, signals that the therapy was not curative. It occurs in 20% to 45% of patients treated surgically for prostate cancer.

"'Recurrence' is a euphemism for disease that was just missed," Strum said.

MR imaging with an added MR spectroscopy sequence (MR/MRS) has been shown in many studies to help localize disease and to plan and monitor treatment. Its sensitivity for cancer detection has ranged from 91% to 95%. This compares with 46% to 61% for MR alone and 63% to 75% for MRS alone (RadioGraphics 2007;27:63-77).

UNCONVENTIONAL IMAGING

Most MR/MRS work is being done in patients with biopsy-proven cancer. Because it is an expensive, time-consuming test, it is unlikely to emerge as a tool in general screening, though it may gain a role as an adjunct screening tool for high-risk men.

UCSF performs about 500 to 600 MR/MRS studies on prostate cancer patients every year and reports that the combined technique has changed patient management in hundreds of cases. Though MR/MRS is mainly done at a handful of academic centers, use has started to trickle out into the community, Kurhanewicz said.

At weekly meetings, UCSF doctors combine clinical with imaging data and plot treatment accordingly. For example, if the disease is truly localized, the patient may need only brachytherapy seeds. In cases of extensive disease, a more systemic therapeutic approach could be warranted.

"If effective, selective imaging helps make these decisions in a more logical fashion. You can save money and improve quality of life," he said.

However, in the prospective multicenter American College of Radiology Imaging Network 6659 trial, accuracy of MR/MRS was similar to that for MR alone for presurgical assessment, according to preliminary results released at the 2006 RSNA meeting. Overall accuracy was in the high 60% range, according to an assessment made by participating centers as of May 2007, when results were still being analyzed.

Such trials indicate that MR/MRS is not yet ready for the population at large, said Dr. Shelley Weiner, chief medical officer at CareCore, a large radiology utilization management company.

"Both endorectal MRI and MRS have great potential. But the techniques have not been adequately defined at this time, and the sensitivities and specificities reported in the literature are very broad. Protocols vary widely," Weiner said. "It has not yet been shown that the technology can be reliably transferred from the investigational setting into everyday practice."

Since the trial was conducted, the MR/MRS software has improved, Kurhanewicz said. MR/MRS is significantly improved on 3T units, which allow a twofold increase in spatial resolution.

Due to the dramatic down-staging that has taken place in prostate cancer detection over the years, the patient population in the ACRIN trial had small amounts of disease, which is harder to pick up with imaging. In contrast, earlier studies reported in the literature featured populations with more extensive disease.

FORTUNE TELLERS

The ACRIN study did not incorporate the use of nomograms, algorithms that allow prediction of risk assessment based on a variety of factors like the Gleason Score and PSA level.

Nomograms have been developed for a variety of purposes, such as for prediction of likelihood that the disease has spread, that treatment will succeed, and that disease will recur after treatment.

The nomogram developed by Memorial Sloan-Kettering's Dr. Michael Kattan, available online at no cost, is particularly well established (www.mskcc.org/mckcc/html/10088.cfm).

Doctors and patients plug various details into the online calculator. If there is a high chance that disease has spread, an endorectal MRI could be appropriate and save costs down the line, Strum said.

A Memorial Sloan-Kettering study of about 570 patients showed significant benefit for use of nomograms in combination with MRI/MRS in predicting nonaggressive disease prior to surgery (Radiology 2007:242[1]:182-188).

Despite advances in nomograms and MRI, initial staging of men with prostate cancer has changed very little in the last 25 years, Strum said. Patients newly diagnosed with prostate cancer continue to get bone and CT scans, even though these are often not appropriate because the studies are not very good at picking up early-stage diseased.

"Although CT continues to be widely used in patients with newly diagnosed prostate cancer, it has virtually no role in prostate cancer detection or staging," Hricak et al said in their prostate imaging review.

MRI could be helpful prior to the new and preferred laparascopic surgical techniques but is often neglected, according to Dr. Mittul Gulati.

"Many urologists are not fully aware of the way MRI could be incorporated into their surgical planning," he said.

While a radiology resident at the University of California, Los Angeles in 2006-2007, Gulati performed a study demonstrating MR/MRS's value prior to surgery in 20 patients (see figure). Results were shown in a poster presentation at the American Roentgen Ray Society meeting in June.

PRESERVING FUNCTION

With RALRP, it is possible to spare the neurovascular bundles (NVBs), which are responsible for erectile function. Accurate MR imaging can provide a presurgical road map of the prostate, showing whether the NVBs are involved and, if so, what needs to be done to ensure a successful surgery.

"The criteria for triaging patients to major surgery for prostate cancer is often based on a blood test and a digital rectal exam-the standard way to risk-stratify patients for decades," said Gulati, now a radiology resident at the Mallinckrodt Institute of Radiology at Washington University.

Advanced techniques are not encouraged in prostate cancer management because they are expensive, even though they will often save money down the line by eliminating unnecessary, expensive treatments, Strum said.

"Physicians should be using and incorporating all the advances. Instead, they are keeping them in a kind of lockbox. In the year 2007, we should be able to say we did whatever we could for the patient," he said.

Prostate MRI utilization has not increased much in recent years. According to Hricak's review, in the centers where it is offered, the number of exams performed per month ranges from one to 150, a "reflection of the lack of consensus about the still-evolving role of this modality."

In the next five to 10 years, researchers will be figuring out the optimal ways to use imaging, Kurhanewicz said. In addition to MRS, diffusion-weighted and dynamic contrast-enhanced sequences can be added to one exam. But it remains to be seen which combinations of sequences are the most valid, cost-effective, and efficient.

"With MRI, validation is in progress. The question is, How do we use these tools in the interim and help patients without fooling ourselves into thinking we know everything? That is the catch-22," he said.

Nanotechnology beckons

There are also some commercial hurdles that have hindered prostate MRI utilization. Four years ago, researchers at Massachusetts General Hospital and University Medical Center in Nijmegen, the Netherlands, found that a lymph node-seeking magnetic nanoparticle agent used in MRI, ferumoxtran-10, was extremely accurate for detecting prostate cancer metastases in the abdomen (NEJM 2003:348[25]:2491-2499).

Although cancer is detected at earlier stages, lymph node metastases are common in high-risk patients, occurring in about 40% of these cases, said Dr. Mukesh Harisinghani, director of abdominal MRI at Mass General.

Boston-based Advanced Magnetics, the agent's U.S. manufacturer, received FDA approval based on certain conditions, including submission of more data.

Later, in December 2006, Guerbet, the company's European partner, submitted a new drug application to the European Medicines Evaluation Agency for approval in pelvic cancers in Europe. The European go-ahead is expected in October or November. Advanced Magnetics is expected to use Guerbet trial data to win approval in the U.S.

"Imaging with nanoparticles will make a huge difference in high-risk men, sparing many from unnecessary surgery," Harisinghani said.

Ms. Hayes is feature editor for Diagnostic Imaging.