Monitoring Cancer's Course: Quantitative PET Treatment Response Assessment

FDG-PET is more accurately predicting cancer treatment response than anatomical imaging, allowing more individualized, successful therapeutic strategies and patient management in lymphomas, breast, lung, gastroesophageal, colorectal and many other cancers. Anatomic imaging criteria have limitations as changes in tumor size frequently occur late during treatment while changes in glucose metabolism provide early response assessments. Quantitative analysis provides new ways of assessing response in specific tumor types, largely guided by the new PERCIST (Positron Emission tomography Response Criteria in Solid Tumors) guidelines.

FDG-PET/CT detects metabolically active tumors. This is especially important in judging response to cancer therapy where post-therapeutic alterations such as scar or necrosis can give the impression of residual tumor when the tumor may in fact have been killed already by the treatment, says John M. Buatti, MD, professor and head, radiation oncology at the University of Iowa Hospitals and Clinics in Iowa City.

Effective early treatment response can be obtained using PET/CT which can be used to make key patient management decisions to continue treatment, halt it or change its course, adds Walter De Wever, MD, PhD, from the Department of Radiology, University Hospitals, Leuven, Belgium.

PET/CT has gained widespread acceptance for the early prediction of response to therapy and outcome in patients with lymphomas, breast, lung, gastroesophageal and colorectal cancers. Studies also document the benefits of FDG-PET for assessing treatment response in patients with other malignancies such as ovarian cancer, uterine cancer, head and neck squamous cell carcinoma, sarcoma, mesothelioma and melanoma.

Sponsored by the The Academy of Molecular Imaging (AMI) in collaboration with the American College of Radiology Imaging Network (ACRIN) the National Oncologic PET Registry (NOPR) was initiated, in 2006, to systematically collect clinical and demographic data on the usefulness and impact of PET and PET/CT in previously cancer types and indications previously not covered by the U.S. Centers for Medicare & Medicaid Services (CMS). The NOPR results published in May 2008 in the Journal of Clinical Oncology reported that PET or PET/CT resulted in a change in intended patient management in 36.5 percent of cancer cases. In a subsequent study, published in December 2008 in the Journal of Nuclear Medicine, the NOPR investigators reported that physicians changed their intended patient management in 38 percent of cases by using PET or PET/CT. The percentage of change to patient management based on PET or PET/CT ranged from 48.7 percent for myeloma to 31.4 percent for non-melanoma skin cancer. In a subsequent study published in Cancer, the same group reported similar impact of PET on managing patients who underwent chemotherapy.

Moving FDG-PET into cancer care guidelines

Of the various medical society guidelines addressing the many different types of tumor-forming cancers, the majority have not yet integrated FDG-PET or FDG-PET/CT as a recommended or optional diagnostic tool as part of disease response evaluation—but this is starting to change. One good example is lymphoma. Lymphoma oncology societies have been leaders in adopting FDG-PET into their tumor response criteria, says Dunphy.

Treating lymphoma

Metastatic Breast Cancer: Before starting a new treatment regimen, FDG-PET visualized widespread metastatic involvement of the pleural lining of the lungs, liver and skeleton in a woman with biopsy-proven metastatic breast cancer. After two cycles of the new treatment, FDG-PET image shows a remarkable reduction in overall tumor burden, including the skeleton. Image source: Mark Dunphy, DO, assistant attending physician in nuclear medicine service at Memorial Sloan-Kettering Cancer Center and assistant professor of radiology at Weill-Cornell Medical College in New York City.

PET using FDG has emerged as a powerful functional imaging tool for staging, restaging, and response assessment of lymphoma. Juweid et al evaluated the impact of integrating PET into the International Workshop Criteria (IWC) criteria in a retrospective study of 54 patients with diffuse large B-cell non-Hodgkin lymphoma (NHL) who had been treated with an anthracycline-based regimen in the July 2005 issue of Journal of Clinical Oncology. PET increased the number of complete response patients, eliminated unconfirmed complete response, and enhanced the ability to discern the difference in progression-free survival between patients experiencing complete response and partial remission. Such findings provided rationale for incorporating PET into revised criteria. The International Harmonization Project has provided new guidelines after incorporating PET for definitions of response in non-Hodgkin and Hodgkin lymphoma. The recommendations for the use of PET or PET/CT are:

• PET is strongly recommended before treatment for patients with routinely FDG-avid, potentially curable lymphomas (diffuse large B-cell lymphoma [DLBCL], Hodgkin lymphoma). For incurable, routinely FDG-avid, indolent, and aggressive histologies (follicular lymphoma and mantle-cell lymphoma), and for most variably FDG-avid lymphomas, the primary end points for clinical trials generally include progression-free survival, event-free survival, and overall survival. PET is not recommended before treatment unless response rate is a major end point of the trial.

• Numerous studies have demonstrated that PET performed after one to four cycles of multi-agent chemotherapy predicts therapeutic outcome; however, until available data demonstrate improvement in results by altering treatment based on this information, this practice should be restricted to clinical trials evaluating PET in this context.
• PET is essential for the post-treatment assessment of DLBCL and Hodgkin lymphoma because a complete response is required for a curative outcome. However, PET is recommended in the other, incurable histologies only if they were PET positive before treatment and if response rate is a primary end point of a clinical study.

Detecting bone metastases

Bone metastasis is an example where the most widely used criteria are based on the anatomic measurement of solid tumors. We most commonly see bone metastases in breast cancer and prostate cancer patients, says Colleen M. Costelloe, MD, associate professor, department of diagnostic radiology, division of diagnostic imaging, University of Texas MD Anderson Cancer Center in Houston. Many therapeutic drug trials take place at MD Anderson, and these trials sometimes offer the last chance for therapy.

Tumor response criteria have been developed to create standard methods of tumor measurements to allow the results of different trials to be compared. The most commonly used criteria are the RECIST (Response Evaluation Criteria in Solid Tumors) 1.1, but it only considers bone metastases with soft-tissue masses less than 10 mm to be measurable disease. Soft-tissue extension from bone metastases is relatively uncommon. Therefore, patients with metastases only to bone can be excluded from the trials. The University of Texas MD Anderson Cancer Center (MDA criteria) specific to bone metastases could potentially take the place of RECIST 1.1 criteria and allow patients with metastases only to their bones to enroll in clinical trials to use study drugs aimed at overcoming their disease, says Costelloe. “The MDA criterion uses radiography, CT and MRI and we hope to update it to include PET/CT,” shares Costelloe.

More quantitative cancer monitoring

Based on the extensive literature now supporting the use of FDG-PET to assess early cancer treatment response as well as the known limitations of anatomic imaging, updated draft PET criteria for solid tumors known as PERCIST were developed and proposed by Richard L. Wahl, MD, professor of radiology and nuclear medicine and director of division of nuclear medicine/PET and colleagues at Johns Hopkins University School of Medicine, Baltimore, last year. Standardization of acquisition and analysis of PET data is essential for comparability of studies across facilities. The criteria provide a framework for more quantitative analysis among both measurements and users and allow exploration of alternative metrics. These may be useful in clinical trials and possibly clinical practice.

PERCIST establishes performance standards for PET scans in a method consistent with the National Cancer Institute recommendations and those of the Netherlands multicenter trial group on well-calibrated and well-maintained scanners. PERCIST is a proposed methodology to make measurements in PET more consistent by using a baseline background SUV within the liver as a way of creating an internal standard of background and a consistent one cubic centimeter area for tumor SUV that would be compared at a response time, says Buatti.

“Anatomic imaging alone using standard WHO, RECIST and RECIST 1.1 criteria have limitations, particularly in assessing the activity of newer cancer therapies that stabilize disease, whereas 18F-FDG PET appears particularly valuable in such cases,” according to the Journal of Nuclear Medicine PERCIST supplement (J Nucl Med. 2009 May;50 Suppl 1:122S-50S; http://jnm.snmjournals.org/cgi/content/short/50/Suppl_1/122S ).

“The proposed PERCIST 1.0 criteria should serve as a starting point for use in clinical trials and in structured quantitative clinical reporting. Undoubtedly, subsequent revisions and enhancements will be required as validation studies are undertaken in varying diseases and treatments.” (visit: http://www.ncbi.nlm.nih.gov/pubmed/19403881)

In PERCIST, response to therapy is assessed as a continuous variable and expressed as a percentage change in standardized uptake value (SUV) corrected for lean body mass (SUL) peak between the pre- and post treatment scans. Briefly, a complete metabolic response is defined as the visual disappearance of all metabolically active tumors.

A partial response via PERCIST is considered more than a 30 percent and a 0.8-unit decline in SUL peak between the most intense lesion before treatment and the most intense lesion after treatment, although not necessarily the same lesion. More than a 30 percent and 0.8-unit increase in SUL peak or new lesions, if confirmed, is classified as progressive disease. A greater than 75 percent increase in total lesion glycolysis is proposed as another metric of progression. “This standardization and a required 30 percent decrement with consistent times make the method important as a standard. This method should apply to all [patients who] have PET avidity and we know some do not.  In those, it is important to use CT based indexes such as RECIST,” Buatti says.

The criteria of a PET/CT system over a stand-alone PET are normally the same. Initially some FDG-uptake could be interpreted as normal on stand-alone PET imaging, but now with the anatomical correlation of CT which is a major advantage or PET/CT, very small lesions visible on CT are made positive on PET/CT even with a slight uptake of FDG, says Wever.

PERCIST in lung cancer patients

“We have adopted PERCIST criteria in one upcoming trial with lung cancer patients for evaluating response to cancer therapy,” shares Mark Dunphy, DO, assistant attending physician in nuclear medicine service at Memorial Sloan-Kettering Cancer Center and assistant professor of radiology at Weill-Cornell Medical College in New York City. Lung cancer patients who consent to participate in this trial will undergo an FDG-PET scan before and after one cycle of conventional therapy (based on the standard-of-care).

The standard-of-care depends on the stage of the lung cancer and varies from surgical resection to adjuvant chemotherapy, adjuvant radiotherapy, concurrent chemoradiotherapy, induction chemoradiotherapy followed by resection, combination chemotherapy and combination chemotherapy with concurrent hyperfractionated radiotherapy with concurrent hyperfractionated radiotherapy.

If the patient’s disease does not appear to be responding by PERCIST criteria, then the patient may be switched to a new experimental therapy. A poor PERCIST response includes, for example, a failure of the tumor SUV scores to decrease significantly.

Dunphy adds that PERCIST criteria are not necessarily prognostic, but provide a formalized approach to deciding what post-treatment changes in FDG-PET scans are real—not merely technical artifacts. In particular, the criteria offer guidance in deciding what changes in the semi-quantitative SUV scores that measure tumor metabolism can be interpreted as a real change. “Just because a patient has a decrease in SUV doesn’t necessarily mean the patient will have a significant clinical benefit from a therapy,” he notes. “You need to have long-term clinical data to justify that inference—and then it truly applies to the particular cancer population undergoing the particular therapy being studied. In general, however, the medical literature has shown that big drops in SUV scores, after treatment, have been associated with better patient outcomes, for a variety of cancer populations and for various therapeutic regimens.”

“To validate FDG-PET as a guide for deciding early after starting a treatment whether or not a patient and his or her oncologist should change the therapeutic regimen, the most rigorous evidence-based approach are randomized therapy trials in which the patient’s early disease-response on FDG-PET is compared to how the patients fare if they stay with the same treatment or are switched to an alternative treatment,” according to Dunphy.

The concept of early treatment response assessments using FDG-PET has been introduced and is increasingly accepted in the medical community. Remarkable progress has been made in establishing FDG PET/CT imaging as the standard of care in cancer imaging. Novel imaging probes might emerge that will allow us to characterize tumors even more comprehensively which then could be used to assess tumor responses to treatment with an even higher accuracy. Stay tuned.

What do Non-FDG tracers offer?

Although FDG is the mainstay for PET, it is not suited for all applications and, in particular, for monitoring the effectiveness of highly specialized therapies. FDG-PET has difficulty in detecting tumors of the brain and urinary tract, while new tracers such as 18F-fluorothymidine (18F-FLT) offer clear visualization of disease in the brain and  bowel, says Mark Dunphy, assistant attending physician in nuclear medicine service at Memorial Sloan-Kettering Cancer Center, and assistant professor of radiology at Weill-Cornell Medical College, New York City. New tracers also are being developed for evaluating tumor response that is more selective for the therapeutic agent involved. For example, FDG detects tumors based on their increased glucose metabolism. Yet certain new forms of cancer agents might not have major or rapid impact on tumor glucose metabolism or may be more “cytostatic”—stabilizing tumors rather than killing tumors. Newer PET agents that detect the specific molecular target or associated biological pathways of these new targeted cancer drugs in tumors might be more relevant to evaluating how well these tumors are responding, he adds. Unlike FDG which now has a huge bulk of clinical data backing it, the number of clinical trials undertaken in non-FDG tracers is usually very limited. Basically, to date, there are no major published PET response criteria for non-FDG tracers because these are mostly considered experimental tracers, with not yet enough clinical experience to validate tumor response criteria. They are not yet available to most medical centers—but that will soon change, Dunphy says. The non-FDG PET tracers that are in the final stages of preclinical development or in the early stages of clinical application for monitoring therapeutic response have been divided into four types—radioactive tracers for DNA synthesis, hypoxia imaging, hormone receptor imaging and amino acid tracers—in a review published by Dunphy and Jason S. Lewis, PhD, professor and vice chair, department of radiology, Memorial Sloan-Kettering Cancer Center in New York City in the Journal of Nuclear Medicine in May 2009. The two leading agents among the radiotracers of DNA synthesis are the thymidine analogs FLT and 18F-fluoro-5-methyl beta-D-arabinofuranosyl-uracil (18F-FMAU). FDG is still the mainstay and FLT will be the next one to be give the regulatory nod, according to Lewis. There is some evidence that FLT can detect tumor proliferation very rapidly within 24 hours, says Dunphy. The earlier you determine whether a tumor is responding at a cellular and biomolecular level, the earlier you will be able to predict whether the tumor is going to respond. In one brain cancer trial, he notes, patients who had a decrease in FLT uptake were better responders to a therapeutic regimen than patients who did not have a decrease in the uptake. Promising agents for PET imaging of hypoxia within tumors include 60/62/64Cu-labeled diacetyl-bis (4N-methylthiosemicarbazone) (60/62/64Cu-ATSM) and 18F-fluoromisonidazole (18F-FMISO). PET imaging of tumor hypoxia might help radiation oncologists in planning radiotherapy—boosting the radiation dose to the  more hypoxic tumors or tumor sub regions, as hypoxic tumors are more resistant to radiation, according to Dunphy. There also are agents for the imaging of tumor expression of androgen and estrogen receptors, such as 16b-18F-Fluoro-5a-Dihydrotestosterone (18F-FDHT) and 16-a-18F-Fluoro-17-b-Estradiol (18F-FES), respectively. Patients with high uptake of 18F-FES also have high expression of the estrogen receptor and seem more likely to respond to hormonal therapy. Similarly, PET agents targeting tumor somatostatin receptors that show higher uptake of the somatostatin tracer predicts a favorable tumor response to somatostatin targeted therapeutic agents, shares Dunphy. Radiolabeled amino acids for PET imaging, such as L-[methyl-11C] methionine (11C-MET), also seem to have advantages over FDG-PET for detecting cerebral gliomas.