Just In... Meeting News

Key news updates and reports from the latest meetings in oncology and hematology.

Monday, July 15, 2013

ONLINE FIRST: Debating the Role of Interim PET Scans in Lymphoma



CHICAGO -- Positron emission tomography (PET) scans have multiple possible uses in lymphoma, including in staging, response evaluation at end of treatment, and in follow-up, but the biggest controversy in lymphoma today is the role of interim PET scans, a topic that was discussed here at a state-of-the-art update at the American Society of Clinical Oncology Annual Meeting.

 “We need early detection of treatment failure -- that is, after a few cycles of chemotherapy or at midpoint of scheduled treatment, which could alter the treatment strategy,” said Andrew D. Zelenetz, MD, PhD, a lymphoma specialist who is now Vice Chair of Medical Informatics at Memorial Sloan-Kettering Cancer Center. “We could alter treatment immediately. If the positive predictive value of interim PET is very high, this justifies the examination of alternative treatment. Why go on? Let’s switch treatment.


“Or we could alter treatment after the end of therapy. If the positive predictive value of interim PET is moderate, slow responders might require closer surveillance during follow-up. We could switch therapy at a later time point when they have disease progression.”


Describing the hurdles of incorporating interim PET for risk-adapted therapy, he said that evaluations must reliably identify patients who will not respond: “We need to demonstrate that there is an effective alternative therapy. The outcome with early use of alternate therapy must be equivalent to or superior to the use of the same therapy at time of treatment failure.”


Interim PET is useful for Hodgkin lymphoma patients. Regardless of the performance status score, “if a PET scan is negative after two cycles, patients do very well; and if the PET scan is positive, they do very poorly,” he said, noting that in early-stage disease, PET 2-positivity is predictive for outcome.


Studies show that a clinician can risk-adapt based on PET scan results. Using PET it is possible to identify a population with good outcome in early-stage Hodgkin lymphoma, but standardization is needed across treatment centers, he said.


“We need longer follow-up to establish the impact of a PET-directed approach on survival and cause of death, as well as further studies in advanced disease.”


Interim PET can make a statistically significant difference in diffuse large B-cell lymphoma (DLBCL) as well. “If a PET scan is positive, that’s bad. But almost half of DLBCL patients still do well in the long term. It’s not a fait accompli that they will do badly if a PET scan is positive,” he said.


Zelenetz’s lymphoma group has conducted two studies of risk adaption for poor-risk DLBCL patients using sequential R-CHOP to ICE therapy combined with interim PET scans. “We found a significant difference in outcome with Deauville criteria in both PET-negative and PET-positive scans,” he said. There was a high false-positive rate, but nearly two-thirds of the patients were disease-free in the long term despite Deauville positivity.”


He noted that if patients with a positive biopsy were removed from the analysis, the PET scan was not predictive. One way to make PET more predictive is to remove patients with primary medial large cell lymphoma, he said: “It is extremely likely that these patients will have a false-positive PET scan, and, therefore, a PET scan is never indicated for them, without further advances.”


PET Scans for Staging

Zelenetz noted that PET scans are valid for staging since virtually all lymphomas are FDG-PET avid. The exception is small lymphocytic lymphoma (SLL), where significantly more disease is seen on contrast-enhanced computed tomography (CT) than with PET.


PET scanning rarely upstages disease, he said. In general, PET scans show more disease in normal-size lymph nodes. Normal-size spleens can also be declared abnormal. The greatest discrepancy is in extranodal sites in the bone and skin. It has been shown that overall, PET scans miss approximately 14 percent of cases of DLBCL, six percent of Hodgkin lymphoma cases, and 40 percent of follicular lymphoma cases.

PET scans do well in staging patients with bone marrow disease. “PET is clearly superior to CT for bone involvement,” he said, adding that bone marrow only (without bone disease) tends to be missed by PET.


He said that FDG-PET imaging can distinguish between patients with indolent and aggressive lymphoma: “There is a substantial overlap in the lower range of SUV [standardized uptake value], but mostly we see indolent lymphoma in the higher range of SUV and aggressive non-Hodgkin lymphoma in the lower range. The likelihood for aggressive disease increases in parallel with increases in SUV. PET results can help guide the site for the optimal biopsy if there is concern about transformation of indolent lymphoma.”


Although PET scans cannot replace biopsy, they can be used to evaluate the response of patients at the end of treatment, Zelenetz added. “The general principle is that if there is persistent FDG uptake after therapy, it indicates a high risk of relapse. A negative PET at the end of therapy indicates good prognosis with regard to progression-free and overall survival [PFS and OS]. This is true for Hodgkin lymphoma, DLBCL, and follicular lymphoma.”


In 2007, standard response criteria of PET scans were integrated into International Workshop Criteria (IWC). “Including PET in the response criteria eliminates complete response-uncertain patients and partial responses that are simply residual masses. There is better definition of favorable prognostic groups if surveillance imaging is eliminated,” he said.


He noted that there is still somewhat limited availability of PET scans outside the U.S., the cost is high, and PET scans have not been validated for all histologies. 


Uniform response criteria would help “to level the playing field” and lead to better interpretation of PET scans, he said. “We need standardized reporting of what is positive and negative, and reproducibility across nuclear medicine physicians and machines.”


The London Deauville criteria, which is a five-point visual inspection scale compared with normal structures, has been proposed. “These criteria are not absolutely applicable,” he said. “In the setting of aggressive disease, with a question of escalation of treatment, positive scans are scored at 4 or 5. In limited-stage disease, with a question of limiting treatment, scores tend to be more conservative, and a score of 3 is positive.” The inclusion of PET scans into the IWC has led to better discrimination between partial and complete responses.


PET Scans in Follow-Up Remains Unproven

The role of PET in follow-up remains unproven and not recommended for routine use, Zelenetz said. “Increasing evidence suggests a limited role for routine imaging in detection of relapse in both DLBCL and Hodgkin lymphoma. No studies have demonstrated superior outcome when relapse is detected by PET versus CT.”


PET is associated with a risk of false-positive FDG-avidity, but that cannot be taken as evidence of relapse, he emphasized. In highly selected cases -- for example, primary bone lymphoma -- FDG-PET may be better for detection of relapse. “In the absence of new data, FDG-PET should not be used for routine monitoring of patients after therapy. There is no role for PET in follow-up except in rare circumstances when a patient might have extranodal disease.”


In summary, Zelenetz said, “Interim PET in Hodgkin lymphoma has been shown to be a powerful predictor of outcome of standard therapy. Deauville visual inspection is useful.” The RAPID trial showed that PET-based risk adaptation can reduce therapy in good-risk patients. “PET-based risk adaptation can alter bad biology -- dose escalation.”


In DLBCL, there is persistent controversy regarding the value of the interim PET. “Many studies show small or no impact. Deauville criteria need further validation in large datasets,” he said.


Bone marrow remains an important tool in staging for DLBCL and follicular lymphoma, but Zelenetz argued that “missing six percent of bone marrow involvement in Hodgkin lymphoma is important, too. I use PET in Hodgkin lymphoma.” PET can also help direct biopsy in cases of suspected transformation of indolent to aggressive disease.


RAPID Results

The moderator of the session, Nancy Bartlett, MD, Professor of Medicine in the Division of Oncology at Washington University in St. Louis, noted that several large European and North American trials are now evaluating interim PET-directed therapy in patients with early-stage Hodgkin lymphoma, with the goal of minimizing chemotherapy cycles as well as avoiding or limiting radiation therapy.


She pointed to the RAPID trial, in which patients with non-bulky stage I-IIa Hodgkin lymphoma received three cycles of ABVD (doxorubicin, bleomycin, vinblastine, and dacarbazine) followed by a PET/CT scan. Patients with a negative scan were randomized to invoIved-field radiation therapy (IFRT) or observation. All patients with a positive interim PET scan received one additional cycle of ABVD and IFRT. Scans were considered positive if the London Deauville visual scale score was 3 or higher.


Of the 602 patients included, one-third had stage I disease and two-thirds had stage II. Three-quarters of the patients had a negative interim PET and were randomly assigned to have IFRT or observation.


At a median follow-up of 49 months from randomization, “PET-positive patients had an excellent three-year PFS of 86 percent and OS of 94 percent,” she said. PET-negative patients randomly assigned to observation had a PFS rate of almost 91 percent compared with 94.5 percent for those who received IFRT, with three-year overall survival rates of 99.5 and 97 percent,  respectively.


For PET-negative patients, the three-year PFS for those who received IFRT was 97 percent compared with about 91 percent for those assigned to observation. Of the seven deaths in the IFRT arm, five occurred in patients who never received radiation.


“Universal application of these trial results will require strict adherence to the Deauville criteria used in the trial,”  Bartlett said. “Outcomes for PET-negative patients treated with three cycles of ABVD with or without radiation therapy were excellent, but because of the lower limit of the confidence interval, the trial does not rule out an advantage in PFS with the use of IFRT.


“Importantly, even the PET-positive patients had an excellent outcome when treated with four cycles of ABVD and IFRT, arguing against altering chemotherapy regimens on the basis of the interim PET.”


SOURCE: Jorge Carrasquillo, NIH