Presented previously as a poster at the 53rd American Society of Hematology Annual Meeting and Exposition; December 10-13, 2011; San Diego, Calif.
Patients with early-stage, nonbulky classic Hodgkin lymphoma (cHL) undergo intensive posttreatment radiologic surveillance despite having a low risk of disease recurrence. The current study attempted to evaluate the risk of disease recurrence and the value of radiologic surveillance in patients treated with the combination of doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) alone who achieved a complete remission (CR) as noted on posttreatment positron emission tomography (PET).
Forty-seven patients who underwent therapy with interim and/or posttreatment PET scans were evaluated for disease recurrence during ≥ 24 months of follow-up. Their presenting characteristics and imaging results were assessed and interpreted in relation to clinical outcome.
All 47 patients were eligible for analysis. The majority of patients were female (35 patients) with a median age of 28 years (range, 17 years-65 years.). The nodular sclerosing subtype was the predominant histology (41 patients). A total of 34 patients were staged with IIA disease, 6 with IA disease, 6 with IIB disease, and 1 with IIEA disease (lung) (according to Cotswolds modification of the Ann Arbor staging system). All patients completed 6 cycles of planned ABVD therapy and achieved a CR. Two had a positive PET scan (1 interim scan and 1 posttreatment scan); both were biopsy-proven sarcoidosis. Two patients developed disease recurrence at 7 months and 24 months, respectively, after negative interim and posttreatment imaging. One case of recurrence was identified through surveillance imaging and the other was identified simultaneously by the patient and surveillance scan. A total of 45 patients experienced a durable CR; 21 had additional unscheduled imaging/workup during surveillance to investigate symptoms or imaging signs of concern.
The goal of treatment in patients with classic Hodgkin lymphoma (cHL) is cure and impressive 5-year overall survival rates of ≥90% have been reported in patients with early-stage disease.1 To minimize late treatment-related effects, research has now turned toward risk stratification and reducing treatment to improve mortality and morbidity without compromising outcome. Accordingly, there has been a shift in the treatment paradigm toward using chemotherapy alone2-4 or reducing the number of chemotherapy cycles and/or the radiation dose/field in combined modality treatment (CMT)5 in patients with early-stage, nonbulky cHL.
Paradoxically, although reductions in therapeutic radiation exposure in patients with cHL are currently being made, the use of diagnostic imaging to evaluate disease has grown with the advent of positron emission tomography (PET) scanning with 18-fluorodeoxyglucose (18-FDG), computed tomography (CT), and radiographs. Although posttherapy imaging is justified to manage patients with early disease recurrence using salvage therapy, there are concerns regarding cumulative radiation exposure and the potential risks.6 This is pertinent in cases of early-stage, nonbulky cHL because patients are usually young with a low risk of disease recurrence.5 Moreover, evidence suggests that most cases of disease recurrence are identified independently of imaging7-11 and the benefit for earlier disease detection is minimal,12 leading some clinicians to call for a reduction in the use of imaging, with European guidelines recommending no routine CT follow-up altogether.13
Based on the Memorial Sloan-Kettering Cancer Center (MSKCC) report demonstrating the equivalence of ABVD with CMT,2 patients with early-stage, nonbulky cHL have been treated, off protocol, with 6 cycles of the combination of doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) alone since 2004. In this retrospective study, we assessed their outcome within the context of radiographic findings throughout treatment and follow-up to identify whether a complete remission (CR) noted on PET scan after ABVD alone can predict a durable remission and obviate the need for further posttherapy follow-up imaging.
MATERIALS AND METHODS
We queried the MSKCC institutional database Web-based data search engine (DAVInCI) to identify patients who were treated with 6 cycles of ABVD without radiotherapeutic consolidation from January 1, 2002 to December 31, 2008. DAVInCI is an MSKCC Web-based application that enables independently run data queries. Patients were eligible if they met the following criteria: early stage (IA to IIB), nonbulky (single lymph node mass measuring <10 cm or a mediastinal mass ≤33% of the maximum transverse transthoracic diameter on a standard posteroanterior chest radiograph at the level of the T5-T6 intervertebral disk) cHL of any subtype; had completed 6 cycles of ABVD successfully at MSKCC (and not at an external facility); received a staging PET scan before treatment; and underwent interim or posttreatment PET restaging and appropriate follow-up (≥24 months posttreatment or until evidence of treatment failure). Pediatric patients or any patients demonstrating CD20-positive cHL (who will likely benefit from rituximab-based therapy14 and have been managed separately at MSKCC15), composite lymphoma, multiple malignancies, human immunodeficiency virus infection, or refractory disease during 6 cycles of ABVD were excluded (Fig. 1).
All scans were performed on PET/CT cameras, including Discovery LS, Discovery ST, and Discovery STE (all manufactured by GE Healthcare, Waukesha, Wis) or Biograph LSO-16 (Siemens Medical Solutions, Malvern, PA), without preference. Patients fasted for 6 hours before undergoing an 18-FDG PET/CT scan; the liberal intake of water was encouraged. Blood glucose was measured at the time of arrival with an institutional cutoff of 200 mg/dL. Approximately 12 to 15 millicuries (mCi) of 18-FDG was injected intravenously. During the subsequent uptake period of 60 to 90 minutes, patients also drank diluted oral contrast medium (diatrizoate meglumine, 2.5% solution, 1000 mL; Bristol-Myers Squibb, New York, NY) while resting quietly in a reclined chair. Low-dose CT (120-140 kilovolt, 80 milliampere) and PET emission images were obtained from the base of the skull to the middle of the thigh. All PET, CT, and fusion images were reviewed on a workstation integrated with a picture archiving and communication system (Volume Viewer 2, AW Suite, version 2.0; GE Healthcare) that allowed multiplanar image reformatting.
All PET scans were reviewed by MSKCC board-certified nuclear medicine personnel before management decisions were made. Each PET scan was reevaluated for the current study by 2 nuclear medicine physicians (H.S. and R.C.L.). Patients were required to have received a pretreatment PET scan (from any institution) to act as a baseline for future studies (to assess 18F-FDG avidity of disease) and for that pretreatment PET scan to have abnormal 18-FDG uptake that was indicative of disease. On-treatment and end-of-treatment scans were performed at MSKCC or a local imaging facility that followed the MSKCC PET scanning protocol, thereby minimizing test variability. Patients who received only gallium scans were excluded.
Using the DAVInCI search results, the medical records of all eligible patients were accessed to ascertain their presenting signs and symptoms, pathology, staging, laboratory values, chemotherapy regimen, follow-up (including imaging) and clinical outcome. This retrospective review was approved by the Institutional Review Board; patient consent was not required. Patient characteristics are outlined in Table 1.
At the time of presentation to MSKCC, all patients received a workup including history and physical examination, blood tests, biopsy review, and CT and PET scans if not already completed at a prior institution. All outside pathologic and radiologic information was submitted and reviewed by MSKCC physicians before treatment planning. All patients were staged according to the Cotswolds modification of the Ann Arbor staging system.16
All eligible patients received ABVD according to MSKCC guidelines (Fig. 2). Doses of chemotherapy were as follows: doxorubicin at a dose of 25 mg/m2, bleomycin at a dose of 10 units/m2, vinblastine at a dose of 6 mg/m2, and dacarbazine at a dose of 375 mg/m2 administered intravenously on days 1 and 15 of a 28-day cycle for a total of 6 cycles. Dose modifications, discontinuance of bleomycin, and growth factor support were as per MSKCC guidelines.
All patients were placed on surveillance and imaged, after treatment, in accordance with standard MSKCC and National Comprehensive Cancer Network (NCCN) guidance, using chest radiographs and CT (1 per year) (Fig. 3). Further imaging (outside of and in addition to this guidance, including CT and PET scans) was requested on an individual basis and in accordance with clinical need.
A total of 337 patients were identified as having received treatment with the ABVD regimen at MSKCC; 47 patients were eligible for the study. Patients were deemed ineligible because of 1) advanced stage of disease, 2) planned consolidative radiotherapy or other treatment, 3) incomplete imaging/medical records, 4) bulky disease, 5) pediatric cases and inadequate follow-up, 6) other medical conditions or recurrent disease after undergoing primary irradiation, or 7) CD20-positive cHL (Fig. 1).
The majority of patients were female (35 patients) and young, with a median age of 28 years (range, 17 years-65 years). Forty-four patients were aged < 45 years. Nodular sclerosing subtype was the predominant histology (41 patients), with 2 patients presenting with mixed cellularity and 1 with lymphocyte-rich disease. The histologic subtype in 3 patients was unspecified. A majority of the patients (34 patients) were staged as having IIA disease, 6 patients were staged as having IA disease, 6 patients were staged as having IIB disease, and 1 patient was staged as having IIEA disease (lung). The majority of patients presented with favorable disease as per NCCN guidelines (34 patients with ≤1 NCCN risk factor). Ten patients had >3 lymph node sites and 1 patient had extranodal disease. Hematologic parameters were predominantly within the normal range: 3 patients were anemic (≤10.5 g/dL), 4 patients had an elevated white blood cell count (>15 × 109/L), and 7 patients had low albumin (<40 g/L). Lactate dehydrogenase was increased (>200 U/L) in 4 patients and the erythrocyte sedimentation rate was increased (≥50 mm/hour) in 12 patients (Table 1).
Response to Therapy
All 47 patients tolerated therapy and achieved a CR. Two patients developed disease recurrence but achieved durable CRs after high-dose chemotherapy and autologous stem cell transplantation.
All patients had abnormal 18-FDG uptake noted on their pretreatment PET scans. Interim restaging was performed after cycle 4B in the majority of patients (40 patients): 39 patients underwent PET scans and 8 patients underwent CT scans only (Table 2). Thirty-eight of the 39 interim PET scans were negative and 1 scan was positive (biopsy-proven sarcoidosis). All 8 patients who underwent interim CT scanning demonstrated a significant response to chemotherapy with either no evidence of, or a marked reduction in, adenopathy.
Table 2. Interim and Post-Cycle 6 Imaging Received
Posttreatment imaging (either PET or CT) took place approximately 4 weeks (median, 2.71 weeks; range, 1 week-12 weeks) after treatment with ABVD. Thirty-three patients had post-ABVD PET scans and 14 underwent CT after a negative interim PET scan (Table 2); 32 of 33 PET scans were negative and 1 was positive (biopsy-proven sarcoid). All 14 patients who underwent CT scans after therapy with ABVD demonstrated either no evidence of disease or stable/shrinking lymph nodes in areas of prior PET-negative disease. One patient had significant thymic hyperplasia. The patient with a previously positive interim PET scan because of sarcoidosis was found to have stable mediastinal and bilateral hilar lymphadenopathy. Both patients who developed disease recurrence after ABVD therapy had negative interim PET scans and a negative posttreatment PET or CT.
Patients were monitored for ≥24 months after post-ABVD restaging (median: 5 months; range, 24 months-95 months) and underwent radiographic surveillance with chest radiographs and/or CT scans according to standard MSKCC practice and NCCN guidelines (Fig. 3).
Two patients developed disease recurrence after treatment. One patient developed disease recurrence at 7 months with progressive lymphadenopathy in previous and new sites of disease, all of which were detected on follow-up CT. There was no palpable lymphadenopathy and the patient was asymptomatic with normal blood work. For the second patient, disease recurrence was confirmed at 24 months after presentation with neck lymphadenopathy, dysphagia, and fatigue. There were no B symptoms and blood work was normal. Concurrent surveillance CT and a subsequent PET scan confirmed enlarging, 18-FDG–avid lymphadenopathy. Although a surveillance scan conducted 6 months prior had demonstrated enlarging lymphadenopathy, a subsequent PET scan was interpreted as negative by an outside institution. Retrospective image review by MSKCC staff, performed as part of the current analysis, revealed abnormal 18-FDG uptake in lymph nodes in the upper mediastinum, which increased on follow-up scan 6 months later.
Remissions: Further Imaging
Twenty-one patients underwent further imaging/evaluation in addition to the standard MSKCC routine surveillance imaging protocol. Reasons included:
Five patients underwent additional PET follow-up to confirm continued CR.
Sixteen patients had abnormal routine surveillance imaging or symptoms that required further evaluation. The majority resolved with observation or antibiotics. Three patients had thymic rebound (identified on surveillance imaging 5 months-7 months after therapy with ABVD and resolving 24 months-58 months after ABVD) and 3 had sarcoidosis (2 of which were diagnosed during treatment with ABVD and 1 of which was diagnosed after treatment during surveillance follow-up).
The remaining 24 patients had an uncomplicated follow-up. All 45 patients remained in CR at the time of last follow-up.
The objective of the current study was to examine whether radiologic surveillance during follow-up in patients with early-stage, nonbulky, CD20-negative cHL who are treated with ABVD alone may be unnecessary because of effective therapy, a predicted low risk of disease recurrence, and low recurrence detection rates with standard imaging. In keeping with the literature, the findings of the current study demonstrate that 45 of the 47 eligible patients achieved durable CRs after ABVD alone. These results reinforce the excellent outcomes with combination chemotherapy reported in other, albeit larger, clinical studies.2-4, 17, 18 We also excluded patients with bulky disease, CD20-positive cHL, and refractory cHL during ABVD therapy from this analysis, thus improving the expectation of a good outcome (Fig. 1).
The standard of care for patients with early-stage, nonbulky cHL has generally been CMT (ABVD plus involved-field radiotherapy) with either 2 or 4 chemotherapy cycles depending on clinical features and initial response. The majority of centers in Europe and the United States have opted for CMT based on randomized trials.5, 19-22 At MSKCC, 6 cycles of ABVD alone has been a treatment option for all appropriate patients with early-stage, nonbulky cHL since our 2004 study comparing ABVD with ABVD plus extended field radiotherapy, which indicated no significant differences in freedom from disease progression.2 Other studies have supported the role of chemotherapy alone in the treatment of patients with early-stage cHL and most recently, reported a survival advantage for this treatment strategy when compared with CMT.3, 4, 17, 18
The cHL recurrence rate ranges from 10% to 15% in its early stages.23 Recurrence is most likely to occur in original or new, adjacent sites,24 with the highest risk of disease recurrence existing in the first year. In the current series, the median follow-up was 55 months (range, 24 months-95 months), and thus was sufficient to capture disease recurrences occurring after therapy with ABVD. The identification of recurrences during posttherapeutic surveillance is important but evidence to support imaging as the key intervention is poor. Patterns of disease recurrence suggest that imaging may be of limited value 2 years after therapy.25 Radford et al demonstrated that the majority of disease recurrences (81%) were identified by the patient, with only 2 cases identified from an abnormal radiograph.9 This has been corroborated by a Canadian study that found 71% of recurrences in patients with HL were identified by the patient.7 It has also been stated that follow-up imaging contributes to earlier disease detection in only 2% of patients.12 These findings have been echoed in subsequent studies.8, 10, 11 Of the 2 patients who developed disease recurrence in the current study, 1 was identified via scheduled imaging and the other was simultaneously identified by the patient and scheduled imaging. It is difficult to determine whether the first case of disease recurrence would have been identified by the patient or on examination, but research suggests that the method of detecting recurrence may have little impact on progression-free or overall survival.7, 26
The US NCCN guidelines for HL recommend imaging on an annual or biannual basis for the first 2 to 5 years of follow-up.27 The 2010 American College of Radiology appropriateness criteria for follow-up also recommends CT imaging for the first 5 years.28 The objective benefit to the patient of surveillance imaging appears to be limited. Although patients can be reassured by a negative imaging scan, they can equally suffer undue distress waiting for results and undergo invasive procedures after a false-positive result. Research has found CT scans to be of little benefit to patients,29, 30 with hypothetical cost-benefit analyses demonstrating CT follow-up to be of minimal survival benefit for patients with all stages of disease and to be associated with a reduction in quality-adjusted life expectancy for those with early-stage disease.29 There are also legitimate concerns regarding the unregulated and cumulative ionizing radiation exposure, particularly in the younger population.7 Nevertheless, it would be understandable if some patients felt a degree of anxiety if they underwent only observation with no surveillance imaging performed at all.
In the current series, 21 patients received further workup and/or imaging (in addition to “standard” MSKCC imaging surveillance practice) because of concerns. At the time of last follow-up, none had developed disease recurrence. Unfortunately, objective assessment of imaging after therapy is compounded by uncertainty regarding residual mediastinal masses, which can represent thymic “rebound”/hyperplasia; necrotic tissue; or active, refractory disease. Therefore, patients are referred for further imaging or biopsy to rule out disease. In the current series, thymic hyperplasia/rebound, sarcoidosis, infection, or a reactive process was found to be largely responsible for abnormal imaging/lymphadenopathy and 3 patients underwent a biopsy to rule out disease recurrence. However, the results of the current study can provide reassurance that in this specific patient population, unusual 18F-FDG avidity, lymphadenopathy, or symptoms are unlikely to represent a case of disease recurrence.
Although it is recognized that interim 18-FDG and end-of-therapy PET scans can provide prognostic information in patients with cHL,31, 32, surveillance patterns vary. Moreover, early disease recurrence within the setting of a negative interim PET scan has been reported.33 However, the pretest probability for disease recurrence in patients with nonbulky cHL is low, and we and others34 have shown that the incidence of recurrence in patients with early-stage nonbulky cHL after a negative posttreatment PET scan is very low. Therefore, in those patients with early-stage, nonbulky cHL, PET scanning could assist clinicians in delineating those who require intensive radiologic surveillance and those who do not. The results of the current study, in conjunction with the growing concerns about radiation from medical imaging, suggest that surveillance imaging in this subset of patients treated with ABVD alone could be ceased altogether.
The results of the current retrospective study indicate that a PET-detected CR noted in patients with early-stage, nonbulky, CD20-negative cHL who are treated with ABVD alone (excluding patients who develop refractory disease during therapy) is predictive of a low risk of disease recurrence and that follow-up imaging studies appear unnecessary. This strategy will reduce cumulative radiation exposure and health care costs in a predominantly young patient population.
Funding support for the current study was courtesy of philanthropy to the Memorial Sloan-Kettering Cancer Center.
CONFLICT OF INTEREST DISCLOSURES
Dr. Hartridge-Lambert received an American Society of Hematology (ASH) Abstract Achievement Award for this work.