CO R R E S P O N D E N C E

To the editor:In the recent article by Willett et al.1 on neoad-juvant therapy with bevacizumab (BV) inpatients with rectal cancer, treatment effectswere ascribed to therapy with the VEGF-spe-cific antibody bevacizumab. However, becausepatients underwent concurrent treatment with5-fluorouracil, we wonder if at least a portion ofthe treatment effects observed at 7 weeks mighthave been mediated by 5-fluorouracil, eitheralone or through a potentially synergistic effectof the combination.

In addition, we regretted the absence ofinformation about the marked increase in 18-fluorodeoxyglucose (FGD) uptake in a projec-tion on the cervical spine region of one patient,as shown by the PET scan 6–7 weeks after theend of neoadjuvant therapy (Fig. 1c in ref. 1).Because this finding was not observed in thispatient’s prior PET examinations, it does notseem appropriate to view it as a mere technicalartifact; it appears highly suggestive of a distantmetastasis that has grown (and consequentlybecome visible by PET) under bevacizumabcombination therapy. Unexpected tumorgrowth at distant sites linked to treatment tar-geted at the primary cancer site, such as havebeen observed in both animal models and can-cer patients2, might be due to higher net sup-pression of growth-inhibiting as compared togrowth-stimulating substances secreted froman individual primary tumor. The PET findingsin this patient could reflect this sort of adverseeffect from bevacizumab–5-fluorouracil com-bination therapy. Therefore, we would like theauthors to comment on the apparent divergentefficacy at the site of the primary tumor as com-pared to distant locations.

1. Willett, C.G. et al. Nat. Med. 10, 145–147 (2004).2. O’Reilly, M.S. et al. Cell 79, 315–328 (1994).

Carsten Goessl1 & Zarko Grozdanovic2

1Novartis Oncology, One Health Plaza, EastHanover, New Jersey, 07936-1080 USA Departmentof Radiology, 2 Charité, Campus BenjaminFranklin, Free University of Berlin, Germanye-mail: carsten.goessl@pharma.novartis.com

C.G is an employee of Novartis, a pharmaceuticalcompany that has antiangiogenic compounds indevelopment.

Willett and Jain reply:The objective of our study was to examine thebiological and clinical effects on rectal cancer ofan initial bevacizumab infusion, before the ini-tiation of radiation therapy and 5-fluorouracil.Six patients with clinical stage T3 or T4 rectalcancer completed the neoadjuvant therapyconsisting of the first bevacizumab infusionfollowed by three 2-week cycles of bevacizumabadministered concurrently with external beamradiation therapy (50.4 Gy in 28 1.8-Gy frac-

tions over 5.5 weeks) and 5-fluorouracil (225mg/m2/24 h by continuous peripheral venousinfusion). Effects of the combined treatment ontumors were evaluated before surgery endo-scopically and by PET scan and perfusion CTscan. Details of the protocol may be found in acorrigendum to our article, printed in thisissue1. The 18-fluorodeoxyglucose uptake inthe neck seen on one representative sagittalimage of the presurgery PET scan (patient 1 inthe original figure) was, in fact, seen on all threePET scans (pretreatment, day 12 and pre-surgery) and was interpreted as upper-limitnormal activity in the thyroid gland. This PETfinding was not judged to represent a metasta-sis, and similarly, PET findings showed nodetectable evidence of metastasis in any of theother five patients (see Fig. 1 below). Althoughthe goal of this study was to examine the bio-logical and clinical response to a single dose ofbevacizumab, the longer-term results of com-bined therapy in six consecutive patients havebeen promising.

Christopher G Willett1, Daniel Duda2,Alan Fischman3 & Rakesh K Jain2

1Department of Radiation Oncology, DukeUniversity Medical Center, Box 3085, Durham,North Carolina 27710, USA. 2Department ofRadiation Oncology, Massachusetts GeneralHospital and Harvard Medical School, Boston,Massachusetts 02114, USA. 3Department ofRadiology, Massachusetts General Hospital andHarvard Medical School, Boston, Massachusetts02114, USA.e-mail willett@radonc.duke.edu

1. Willett, C.G. et al. Nat. Med. 10, 649 (2004).

PET concerns in bevacizumab treatment

NATURE MEDICINE VOLUME 10 | NUMBER 6 | JUNE 2004 561

Pre-Tx Day 12 Presurgery

Patient #2

Patient #3

Patient #4

Patient #5

Patient #6

Figure 1 Effect of treatment on tumors in patientswho completed the entire combined treatmentregimen. Tumor FDG uptake before treatment (pre-Tx), 12 d after BV treatment and 6–7 weeks aftercompletion of all neoadjuvant therapy (presurgery).Sagittal projections of FDG-PET scans for patient2–5 are shown. Tumor is outlined in box, posterior tobladder. On day 12 after BV infusion, the follow-upPET scans showed no change in tumor FDG uptakein five patients and a 40% decrease in patient 3. Sixweeks after completion of the BV, radiation therapyand chemotherapy regimen (presurgery), follow-upPET scans showed decreased tumor FDG uptake ascompared with pretreatment values in patients 1, 2,3, 4 and 6. Tumor FDG uptake in patient 5 was low,and was similar before and at the end of therapy.

©20

04 N

atur

e P

ublis

hing

Gro

up

http

://w

ww

.nat

ure.

com

/nat

urem

edic

ine