Positron emission tomography/computed tomography surveillance in patients with lymphoma: a fox hunt?

Despite improvements in survival rates, relapses after first-line therapy can occur in 20–50% of patients with advanced-stage Hodgkin’s lymphoma (HL) or diffuse large-B-cell lymphoma (DLBCL).1,2 In both diseases, treatment failures are usually observed within 3 years of completion of treatment with the majority of relapses occurring in the first 12 months for HL3,4 and 18 months for DLBCL.4 However, no consensus exists on an optimal surveillance strategy to determine a preclinical relapse after first remission, although routine [18F]fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) imaging has become a standard practice in many centers. The benefits of monitoring for recurrence depend on the probability of relapse in the population being tested as well as the sensitivity, specificity and the frequency of the test.5 The prevalence of relapse in both HL and DLBCL is rare, with reportedly only one relapse per 68 visits in HL6 and per 40–45 visits for patients with aggressive non-Hodgkin’s lymphoma (NHL)5 based on routine CT scans. While PET/CT has not been closely investigated in a surveillance setting, it has become clear that its accuracy is superior to that of CT imaging. According to results from a meta-analysis, FDG-PET predicts disease relapse with a sensitivity and specificity of 50–100% and 67–100%, respectively, for HL and 33–77% and 82–100%, respectively, for NHL, irrespective of the association of a residual mass on CT.7 Even with its relatively high sensitivity, the risks and benefits of routine surveillance PET/CT imaging remain controversial, mainly, because of its cost, radiation burden and the high rate of false-positive results (30–80%)3,8–12 with potential consequences of overtreatment. In this issue of the Journal, El-Galaly et al. report the value of surveillance PET/CT in a retrospective cohort of 161 HL patients who achieved a complete or partial remission after first-line treatment.12 During a median follow-up of 34 months 14% of patients experienced a relapse. With an average of 1.9 PET/CT per patient, the positive predictive value (PPV) of routine PET/CT and clinically indicated PET/CT was 22% and 37%, respectively (P=0.02). However, in a subset of high-risk patients (with extranodal disease, a positive PET result at interim or therapy completion) the PPV increased to 36% while in those without risk factors the PPV was only 5%. Consequently, the authors concluded that the routine use of surveillance PET in HL patients entering complete remission after first-line treatment should be reserved for high-risk patients. These results were in line with those of Petrausch et al. who suggested that monitoring may be worthwhile in high-risk DLBCL populations.11

The weaknesses of the existing surveillance PET studies include retrospective design, the paucity of prospective data in distinct risk categories, non-standardized interpretation of PET/CT, and the lack of randomized multicenter setting, while the strength is that all data were obtained after achievement of first complete remission in both HL and DLBCL. In the study by El-Galaly et al., the major limitation is the retrospective data procurement from existing reports from a period spanning 10 years without using up-to-date PET scanners, standard protocols, or interpretation criteria.12 Notwithstanding major steps taken towards standardizing PET readings in lymphoma,13,14 the reading schemes still vary across centers which may lead to more false positive findings than would be otherwise obtained.

One can readily deduce that frequent PET/CT monitoring is not justifiable for low-risk HL or DLBCL. However, based on the premise that treatment at relapse is more likely to be effective when the disease is in a preclinical stage with a small tumor burden,15–17 routine surveillance imaging for patients in their first complete remission could be theoretically justified.

A multitude of variables should be considered in order to increase the benefits of surveillance PET/CT imaging in the proper population of patients.

Figure 1.Patient with stage IIB abdominal HL (unfavorable) referred for a routine follow-up 6 months after completion of therapy, with no clinical symptoms. Post-completion PET/CT study (upper panel) shows a mass in the right common iliac region (arrows) with no appreciable FDG uptake consistent with treated lymphoma. However, the 6-month follow-up PET/CT study (lower panel) shows a focus of FDG uptake (arrows) within the mass, consistent with recurrent lymphoma.

In conclusion, the preponderance of evidence suggests that in the majority of cases surveillance shortens progression-free survival without translating into a prolonged overall survival. One can confidently state that given both economic and psychological impacts on the patient, routine imaging after first complete remission is not indicated in patients with a low-risk profile including favorable interim response profile. Nonetheless, the debate on the role of surveillance FDG-PET cannot be complete without discrete data on patients with high-risk HL and DLBCL. Although the low PPV of FDG-PET raises important questions about this latter’s clinical value in identifying patients for immediate salvage treatment, the, thus far, unproven survival benefit of such an approach should be addressed in a well-designed prospective, multicenter study employing standardized interpretation criteria to address the dilemma in the high-risk subgroup.

Footnotes

• Andrea Gallamini is Director of the Hematology Department in S. Croce Hospital, Cuneo (Italy) and Professor in Haematology in Nice University (France). He pioneered the use of interim-PET in Hodgkin’s lymphoma to predict treatment outcome, chaired several International workshops on interim-PET in lymphoma, and is now head of the imaging commission of FIL (Fondazione Italiana Linfomi). Lale Kostakoglu is Professor of Radiology and Director of PET/CT Oncology and Research at Mount Sinai Medical Center in New York. She is the chair of the Lymphoma Working Group for the American College of Radiology Imaging Network (ACRIN). She is also a member of the imaging subcommittee of the Cancer and Leukemia Group B (CALGB) (a.k.a ALLIANCE) and a liaison to the lymphoma subcommittee of the same cooperative cancer research group.
• Financial and other disclosures provided by the author using the ICMJE (www.icmje.org) Uniform Format for Disclosure of Competing Interests are available with the full text of this paper at www.haematologica.org.