Assessment of metabolic activity by PET-CT with F-18-FDG in patients with T-cell lymphoma


18F-fluorodeoxyglucose positron emission tomography-computed tomography (PET/CT) has become widely recognized as a suitable diagnostic tool in the staging (Jerusalem et al, 2001), follow-up and tumour response assessment of lymphoma patients (Mikhaeel et al, 2005) and a prognostic role has been reported (Haioun et al, 2005). PET/CT offers the advantage of functional tissue characterization, which is mostly independent from morphological criteria. Different 18F-fluorodeoxyglucose (FDG) uptake among different subtypes of lymphoma has been reported (Schoder et al, 2005), which could be explained by the complex histological profile. The evidence that the degree of FDG accumulation correlates well with the percentage of proliferative cells in the biopsied samples rather than with the histological grade has been reported mainly in B-cell non-Hodgkin lymphoma (NHL) (Leskinen-Kallio et al, 1991). Conversely, only few data have identified the amount of tracer uptake and the role of PET/CT in staging T-cell neoplasms that usually exhibit aggressive clinical behaviour (Tang et al, 2009).

We evaluated the results from the semiquantitative assessment of a staging FDG PET/CT in 20 patients with histologically proven T-cell lymphoma (TL), comparing the findings to the Ki-67 proliferation index and to those obtained in patients presenting indolent- or aggressive B-cell NHL (B-NHL). All patients signed informed consent in accordance with the Declaration of Helsinki.

PET and CT were carried out 75–90 min after intravenous administration of 444–555 MBq of FDG. Focal or diffuse FDG uptake above background in a location mismatched with normal anatomy or physiology was interpreted as abnormal and indicative of a lymphoma lesion. The maximum standardized uptake values, body weight corrected (SUVmax), were determined on PET scans.

Patient characteristics of each group (study and controls) are summarized in Table I. TL and indolent B-NHL had 3·8 ± 2 and 4·7 ± 2 sites involved respectively (P = not significant. ns), whereas extra-nodal aggressive B-NHL had 1·5 ± 0·4 sites (P < 0·05 vs. TL and vs. indolent B-NHL).

Table I.   Patients characteristics.
HistologyPatients (N)Sex (F/M)Mean age, years (range)Number of locations at PET-CT (range)
  1. MF, mycosis fungoides; PTCL-U, peripheral T-cell lymphoma unspecified; ENKL, extra-nodal natural killer/T-cell lymphoma, nasal type; ALCL, anaplastic large cell lymphoma; AILT, angioimmunoblastic T-cell lymphoma; NHL, non-Hodgkin lymphoma; FL, follicular lymphoma; EMZL, extranodal marginal zone B-cell lymphoma; DLBCL, diffuse large B-cell lymphoma; LBCL, large B-cell lymphoma.

T-cell lymphomas
 MF30/362 (52–79)2–64–5
 PTCL-U93/656 (20–76)1–72–10
 ALCL63/349 (21–68)1–12–4
Indolent B-NHL
 FL186/1269 (31–75)1–51–14
 EMZL21/151 (45–58)1–20
Aggressive B-NHL
 DLBCL176/1161 (51–78)1–2
 EMZL Blastic20/248 (21–75)1–1
 Intravascular LBCL10/1401

Mean Ki-67 was 64% ± 15% in patients with TL (range: 30–80%). No relationship between individual mean SUVmax and Ki-67 values was observed (r = 0·1; P = 0·6 [ns]).

Overall SUVmax in patients with TL was lower when compared to either indolent or extra-nodal high grade B-NHL (6·1 ± 4 vs. 9·1 ± 6, and vs. 8·9 ± 5 respectively; P < 0·01).

When the SUVmax was assessed only in the extra-nodal locations, the TL group still presented lower values than the other groups (5·3 ± 3 vs. 9·5 ± 6 and 8·9 ± 5 respectively; P < 0·05) (Fig 1A).

Figure 1.

 Extra-nodal (A) and nodal (B) maximum standardized uptake value in patients with T-cell lymphoma (TL) and indolent or aggressive B-cell non-Hodgkin lymphoma (B-NHL). *P < 0·05 vs. indolent B-NHL; †P < 0·01 vs. aggressive B-NHL.

Finally, the SUVmax in nodal locations of patients with TL was lower than in nodal indolent B-NHL (6·4 ± 3 vs. 8·8 ± 6; P < 0·05) (Fig 1B). Within the TL or indolent B-NHL group there was no difference in SUVmax values between extra-nodal and nodal sites. No differences were found between SUVmax values of indolent B-NHL and extra-nodal aggressive B-NHL.

The present study found that patients with TL had a significant but low metabolic activity at PET scan, regardless of a considerable tumour burden, an aggressive clinical behaviour and high values of Ki-67. These patients had a lower uptake index in the involved regions than sex- and age-matched patients presenting either indolent or aggressive B-NHL.

The unfavourable prognosis of T-cell neoplasms emphasizes the need for more effective diagnostic tools in order to implement beneficial therapeutic strategies.

Schoder et al (2005) reported that the intensity of FDG uptake was generally lower in indolent, compared to aggressive, NHL and that the likelihood for aggressive disease increased in parallel with increases in SUV. This is not the case for TL, in which mechanisms other than those sustaining aggressiveness should be considered.

In our setting, for instance, mean SUVmax of lesions was constantly low, though significant, for involved areas, even though TL patients showed high values of Ki-67, which is an established marker of cell proliferation.

Some authors (Kuo et al, 2008) focused on the potential of PET/CT for improving the staging and the monitoring of response to therapy in TL despite that this method was previously indicated to be less sensitive for the detection of extra-nasal sites. Our TL patients showed a distinctive involvement of extra-nasal sites, mostly cutaneous, visceral or with bone marrow locations and some patients had nodal involvement. Nevertheless, the degree of tracer uptake between extra-nodal and nodal locations of TL patients was not different. Irrespective of locations involved, mechanisms of metabolic deregulation during the genesis of TL and cell transformation might be more complex than merely meeting the cellular requirements of growth and proliferation.

Although clear evidence is lacking at this time, biological differences among specific pathological subtypes of lymphoma can be identified in determining different degrees of FDG uptake. Higher activity of glucose-6-phosphatase might determine a lower FDG uptake as in some histological subtypes of Hodgkin lymphoma (Pangalis & Tsavaris, 1986). There is evidence that some TL cells mirror, histopathologically, mononuclear cells or even the typical Sternberg-Reed cells of Hodgkin lymphoma (Whitcomb et al, 1983). Finally, the expression of the membrane glucose transporter, Glut-1 (Aloj et al, 1999) or the role of lectins that preferentially bind to carbohydrates inside and outside the cells, should also be considered.

TL showed lower FDG uptake in extra-nodal and nodal locations as compared to those of indolent B-NHL. In addition, SUV values obtained at the extra-nodal level in TL were compared with those of patients with extra-nodal aggressive lymphoma; once more the degree of FDG uptake in TL was significantly lower, confirming that reasons other than aggressiveness are implicated in cellular tracer accumulation.

Our findings suggest a careful quantitative evaluation of FDG PET images in such diseases, as higher proliferation does not always indicate higher metabolic activity.

The accurate staging of TL is essential for providing the patient with appropriate prognostic information and for guiding the clinician in selecting the treatment option according to the tumour metabolic activity, which does not seems to be directly related to the proliferation. As a result, alternative drugs, such as receptor-selective, immune-modulating or apoptosis-inducing agents, rather than those disrupting cancer growth by direct cytotoxic effect could be preferred.