The Promise of PET/CT in Oncology

	Screenshot courtesy of J Nucl Med 2011;52:48–55.

While PET/CT is commonly and very successfully used for the staging and follow-up of cancers, researchers are seeking ways to make the modality more sensitive and specific by using targeted radiotracers and refining scanning techniques.

Research paradox

At last year's SNM meeting, there were 362 oncology abstracts (263 diagnostic and 99 therapeutic). "These abstracts, however, represent a research paradox in oncology. We had 278 abstracts on PET/CT and only 19 on SPECT/CT, although the latter is the dominant procedure," opined Steven M. Larson, MD, in an essay of the meeting highlights (J Nucl Med 2010; 51[9]:19N-25N).

"I do not know exactly what to make of this—perhaps there is so much promise in PET, and more people are focusing on that," wrote Larson, chief of nuclear medicine service at Memorial Sloan-Kettering Cancer Center in New York City.

Larson noted that a major avenue of research centered on correlating imaging probes such as FDG with cancer phenotypes. Phenotypic elements include familiar ones such as tissue invasion, metastasis, evading apoptosis and sustained angiogenesis, but also "many ways in which the cancer cell reacts to protect itself against stress," including metabolic, oxidative, proteotoxic, mitotic and DNA damage stress.

One example of an emerging agent noted at the meeting is Aposense, an 18F radiolabeled tracer to image apoptosis in tumors following treatment. "We have long wanted a good agent for imaging apoptosis," Larson wrote, adding that he is "quite intrigued by the fact that the Aposense group has developed a methodology that is a voxel-by-voxel analysis, similar to principal component analysis in MR imaging, which looks at change and direction of change in individual voxels."

While PET/CT is the dominant modality today for cancer staging and follow-up, single-photon imaging—which includes planar gamma camera and SPECT imaging—is commonly used to stage and follow-up skeletal metastatic disease, says Michael Graham, MD, director of nuclear medicine at the University of Iowa Hospitals and Clinics in Iowa City and former SNM president.

Planar imaging is the most common way to perform bone scans, but SPECT is typically used when the cancer activity is deeper and more complex, such as in the spine and pelvis, Graham says. The majority of skeletal metastatic disease comes from patients with prostate, breast and lung cancers.

While PET holds much promise in oncology, Larson challenged imaging physicists to continue working to quantitate SPECT. "This is apparently a very big challenge, because SPECT is now far behind PET in this regard," he said.

Customizing PET

Figure 1. Baseline (top) and week three (bottom) scans of patient with a response to antiangiogenic treatment on all imaging modalities: CT (A), 18F-FDG (B), H2 15O PET (C), and dynamic contrast-enhanced MRI (D). Source: J Nucl Med 2011;52:48–55

FDG is the most common radiotracer used for clinical PET imaging. It is a glucose analog and is taken up into tissues that are actively metabolizing glucose. Tumors utilize glucose in an inefficient manner and, therefore, have to take in more of it. The uptake of FDG in tumors is nicely imaged with PET; however, there are some exceptions.

For example, melanoma, which accounts for up to 90 percent of all deaths caused by skin cancer, has a high metastatic potential if the primary lesion shows any number of adverse features. "However, the sensitivity of 18F-FDG PET is not sufficient in the initial diagnosis of metastases in regional lymph nodes, for which sentinel node biopsy preceded by lymphoscintigraphy is the gold standard," wrote Heikki Minn, MD, and Pia Vihinen, MD, from the department of oncology and radiotherapy at Turku PET Centre in Turku, Finland, in an invited commentary (J Nucl Med 2011;52:5-6).

Researchers are experimenting with radiotracers that relate to specific characteristics of melanoma cells in an effort to better combat recurrence or metastatic disease. One such PET probe is 18F-MEL050, which, in an experimental model, was found to be sensitive and specific in the identification of regional lymph node metastases (J Nucl Med 2011;52:11A-12A).

"The path from original idea—melanin-affine antibodies and peptides to a diagnostic radiopharmaceutical—is long and very similar to that of novel drugs," says Minn. "I regard 18F-MEL050 as one of the most promising compounds since it passed pre-clinical validation and showed, in my opinion, great potential, which now needs to be proven clinically in prospective trials in melanoma patients."

Compared with FDG, the advantage of 18F-MEL050 is both its high specificity and sensitivity and, therefore, it should be able to help both diagnosis and appropriate management of melanoma patients and likely reduce costs associated with overtreatment, Minn says.

While the path to approval and routine clinical use of this or another similar highly specific melanoma probe will probably take three to four years, the potential clinical implications of such a tracer are easy to imagine, he says. Being able to detect smaller metastases than FDG, the new probe would better aid surgery and focused radiotherapy, "which, in the case of melanoma, may best be delivered with conformal beams and higher-than-standard daily fractions." Minn suggests that 18F-MEL050 would improve metastatic staging in the lungs and brain, which would help in the overall choice between different treatment modalities. The melanoma-specific probe also would be better than FDG in imaging metastases of no known primary site.

In addition, "concurrent perilesional and systemic injection of 18F-MEL050 might become the first imaging test preceding sentinel node biopsy and replacing most conventional imaging studies," Minn and Vihinen wrote.

Size vs. metabolism

Figure 2. Transverse slices showing lung lesion (arrows) with 3.5:1 contrast in patient with normal BMI of 19 (top) and liver lesion (arrows) with 2:1 contrast in patient with high BMI of 42 (bottom). Source: J Nucl Med 2011;52:347-353

The rise of antiangiogenic agents for treating tumors also gives rise to the need to find the best way to measure their effect. Tumor shrinkage may not be apparent following treatment, even as metabolic activity within the tumor has undergone significant changes. Molecular imaging is the cornerstone for treatment response imaging.

Adrianus J. de Langen, MD, and colleagues from VU University Medical Center in Amsterdam, the Netherlands, used FDG PET, along with CT, H2 15O PET and dynamic contrast-enhanced MRI, to derive measurements on tumor size, glucose metabolism, perfusion, and microvascular permeability in patients with advanced non-small cell lung cancer treated with bevacizumab and erlotinib (J Nucl Med 2011;52:48–55).

Erlotinib inhibits the function of the epidermal growth factor receptor, and bevacizumab targets circulating vascular endothelial growth factor. Their supposed effects are a decrease in vascular permeability, microvascular density and cell density, according to the study.

As expected, molecular imaging was superior to static imaging with CT in terms of showing early response to treatment. Researchers found that regardless of whether tumor size changed after three weeks, patients with a more than 20 percent reduction in tumor glucose metabolism had a favorable progression-free survival time. In addition, H2 15O PET imaging showed a major flow reduction irrespective of size change and MRI parameters indicative of treatment success also were seen.

"Today, most trials rely on size criteria (RECIST) to base decisions on whether to discontinue drug therapy. However, it can be questioned whether this approach is applicable for trials with targeted agents, looking at our and others' results," researchers wrote.

While the data are promising, more research needs to be done. "Validation of biomarkers predictive for response in oncology is a step-wise process," says senior author Otto S. Hoekstra, MD. "The present study is our first experience in combining PET and physiological MRI information. Such data-driven results require validation. With PET and new interventions, one further needs to reconfirm the association between signal change and clinical outcomes."

Hoekstra says that the use of a PET/MRI scanner would be perfect in this scenario. "The ideal situation would be to measure PET and MR signals simultaneously. With PET, one can only measure one biological process at a time."

In the current study, however, Hoekstra et al measured perfusion and glucose metabolism with PET, but that required two episodes of scanning. Since the half-life of 15O-water is only two minutes, it is technically feasible to combine FDG and 15O-water in a single patient session, he says, but one needs an on-site cyclotron, limiting the applicability of this approach in daily practice.



"With a PET/MRI scanner, besides adding enhanced soft-tissue contrast from structural MR, physiological MR should add functional information—at a higher spatial resolution than PET can offer—to PET's molecular specificity and exquisite sensitivity to get a clear picture of the biological tumor profile," Hoekstra says (see Fig. 1).

The most common cancer imaged with PET is lymphoma, followed closely by lung cancer, says Graham. "More than 50 percent of lymphoma patients receive at least one PET scan during their treatment."

Regarding diffuse large B-cell lymphoma, FDG-PET has prognostic value when performed at the completion of initial chemotherapy and also may be predictive of outcomes when performed during the treatment course. However, robust prospective studies and standardization of interpretation are lacking, according to Amanda Cashen, MD, assistant professor of medicine in the division of leukemia and stem cell transplantation at Washington University School of Medicine in St. Louis, and colleagues (J Nucl Med 2011;52:386–392).

In that regard, Cashen and colleagues conducted a prospective study of 50 patients with advanced-stage diffuse large B-cell lymphoma who were treated with standard rituximab, cyclophosphamide, doxorubicin, vincristine and prednisone chemotherapy. Researchers performed FDG PET/CT after cycle 2 or 3 and at the end of therapy. The PET/CT scans were interpreted according to the International Harmonization Project (IHP) for Response Criteria in Lymphoma. Cashen and colleagues found that the IHP criteria applied to interim scans did not distinguish prognostic groups whose outcomes were sufficiently divergent to warrant a change in therapy.

"The IHP criteria are too stringent for the interpretation of interim PET in diffuse large B-cell lymphoma," Cashen says. "Perhaps a percent or absolute decrease in SUV [standard uptake value], SUV from baseline, SUV greater than a certain level, or FDG uptake greater than liver contrast would be better thresholds to distinguish positive and negative interim PET scans. We explored some of these criteria in our study sample but could not indentify other criteria that predicted relapse. However, larger samples, with consistent use of baseline PET, may find that other criteria are predictive."

Cashen suggests that post-cycle 2 interim PET may be too early in the treatment to identify patients who are having an inadequate response to therapy. "Patients may have residual FDG-avid disease at that time, but then achieve eradication of lymphoma with additional cycles of treatment. Also, at that early time point, PET may be detecting inflammatory changes, not residual disease."

In fact, Emmanuel Itti, MD, from the department of nuclear medicine at the H. Mondor Hospital in Créteil, France, and colleagues examined the same phenomenon and concluded, "When assessing early response, particularly in risk-adapted therapeutic trials, it seems preferable to refer to a background tissue (liver) with a higher level of uptake than that of current international criteria which were designed for end-of-treatment evaluation" (J Nucl Med 2010;51:1857–1862).

Time-of-flight advantages: lung cancer

Figure 3. Patient with stage T2 N0 M0 right hypopharyngeal head and neck squamous cell carcinomas treated with concomitant chemotherapy and radiotherapy. Patient was imaged with intravenous contrast CT, MRI (fat-saturated [FS] T2-weighted sequence), and 18F-FDG PET before treatment and at end of week three (30 Gy). Primary tumor shrinkage was observed with all imaging modalities but was more dramatic with 18F-FDG PET. Source: J Nucl Med 2011;52:331–33

Conventional PET scans create images by detecting gamma rays produced by radioisotopes. Although these conventional scans track where the gamma rays go, they don't consider the time it takes for each gamma ray to reach the detector. Time-of-flight (TOF) PET scans, however, take into account the travel time, which results in improved image signal-to-noise ratios.

Georges El Fakhri, MD, from the division of nuclear medicine and molecular imaging at the Massachusetts General Hospital in Boston, and colleagues recently—and for the first time—demonstrated quantitatively that TOF PET can improve cancer detection in lung and liver lesions (J Nucl Med 2011;52:347-353).

Researchers implemented a modeling technique that used whole-body patient data, rather than simulated data or measured data with physical phantoms. They found the greatest gains in lesion detection were achieved in the shortest-acquisition studies and in the subjects with a BMI of 30 or more. Also of note, the greatest gain in performance was achieved at the lowest lesion contrast and the smallest gain in performance at the highest lesion contrast (see Fig. 2).

"Imagers may be able to take advantage of the gain achieved with TOF PET to reduce scanning time, therefore increasing patient comfort and minimizing patient motion, as well as reducing the amount of the radiopharmaceutical," says senior author Joel S. Karp, MD, from the department of radiology at the University of Pennsylvania in Philadelphia.

Karp says the benefit of TOF PET is characterized in terms of lesion contrast and lesion size, rather than the biology of the cancer. "Therefore, the TOF benefit is potentially more significant for cancers that can be treated if lesions are detected earlier, when they are small and/or have relatively low uptake, or low contrast, or perhaps in catching small lesions that have metastasized from the known, primary location."

Much can happen in four years

Four years ago, Grégoire et al from Catholic University in Brussels, Belgium, wrote that FDG-PET has value for the selection of target volumes in radiation oncology, particularly in non–small cell lung cancers and esophageal tumors. However, they bemoaned the lack of convincing data for head and neck squamous cell carcinomas (J Nucl Med 2007;48:68S–77S). This year, Grégoire and Chiti published a "review of state-of-the-art therapeutic management of patients with head and neck squamous cell carcinomas with all available data justifying the use of PET with 18F-FDG" (J Nucl Med 2011;52:331–334).

While the researchers did not find convincing evidence to recommend FDG-PET in some indications for head and neck cancers (primary tumor assessment, neck lymph nodes), they noted that the use of FDG-PET "has become increasingly popular in radiotherapy planning and has shown promise in various head and neck cancers" (see Fig. 3).

"Although FDG-PET seems a good candidate for radiotherapy planning, its optimal use is not a trivial task," they wrote. "The physics of PET, including poor statistics and poor spatial resolution, result in highly noisy and blurred images that may severely affect accurate determination of the volume and shape of tumor."

They also noted that there are adequate segmentation tools that enhance the promise of PET in radiotherapy planning because it provides smaller, more accurate and reproducible gross tumor volumes than does CT or MRI.

"There's a lot of research being conducted regarding the use of targeted PET probes and finding better ways to show efficacy in various tumors and settings," says Graham. "The field will continue to explore the utilization of PET/CT for pre-operative planning, response to therapy and longer-term follow-up. We will continue to refine techniques and protocols in an effort to optimize the value of molecular imaging in oncology."