TO THE EDITOR:
We read with great interest the paper from Lambert et al. (1) about the treatment of hepatocellular cancer using a mean activity of 3.6 GBq of 188Re-labeled 4-hexadecyl-1,2,9,9-tetramethyl-4,7-diaza-1,10-decanethiol/lipiodol (188Re-HDD/lipiodol) in 11 patients.
We would like to comment on the relatively high urinary excretion rate that was reported in this paper. The authors described a urinary excretion rate of 44.1% ± 11.7% of administered 188Re-HDD/lipiodol within 72 h, which is ineffectively high and comparable to the treatment of bone metastases using radioactive labeled diphosphonates (2). This high urinary excretion rate corresponds to an effective half-life of 14.3 ± 0.9 h in the whole body.
For the purpose of radioembolization, 188Re-human serum albumin (HSA) microspheres were developed and the labeling procedure was improved (3,4). High stability of the radiopharmaceutical could be demonstrated in in vitro studies. The first human applications, in 8 patients (14 treatment sessions) with hepatocellular cancer or liver metastases of colon cancer, revealed a low urinary excretion rate for 188Re-HSA microspheres (8.5% ± 3.6% of administered activity within 96 h). This rate resulted in an effective half-life of 15.7 h in the whole body.
188Re-HDD/lipiodol has some other disadvantages, compared with 188Re-HSA microspheres. Lambert et al. (1) reported a total radiochemical yield of 53% ± 4.5% for 188Re-HDD/lipiodol. In our studies was observed a high radiochemical yield, greater than 95% in vitro, for 188Re-HSA microspheres. 188Re-HDD/lipiodol also showed a relatively high radiolysis (5). This fact was reflected by the instant thin-layer chromatography analysis performed by Lambert et al. (1), which found a high rate of 188Re-perrhenate in urine. In animal studies using 188Re-HSA microspheres (3), an in vivo stability > 90% and an in vitro stability > 88% were observed (investigated in human plasma, blood, and saline for 30 h) (4). Another interesting radiopharmaceutical is 90Y-glass microspheres, but urinary excretion rates were not described (6)
Lipiodol, as an emulsion of iodized ethyl esters of fatty acid of poppy-seed oil, is more a “chemical” embolization agent with a high viscosity (7). Intraarterial injected lipiodol flowed retrograde into the portal venules through hepatic sinusoids and flowed antegrade through the peribiliary vascular plexus (8). This fact indicates only a weak fixation of this agent in the tumor capillary. In contrast, HSA microspheres are “physical” embolization agents, which embolized the capillary vascular plexus. The mean particle diameter of about 25 μm (4) and the uniform size are optimal for embolization (4), and this radiopharmaceutical is widely used in perfusion scanning of the lung.
In view of the relatively high urinary excretion rate, low radiochemical yield, and weak stability of 188Re-HDD/lipiodol, we propose the use of 188Re-HSA microspheres for the radionuclide treatment of hepatocellular cancer and liver metastases to increase the tumor dose and reduce unnecessary radiation exposure to patients.
REFERENCES
REPLY:
My coauthors and I thank Drs. Liepe and Kotzerke for their interesting comments on our paper (1). Urinary excretion of 188Re is 60% ± 12% 48 h after injection of 188Re-hydroxyethylidene diphosphonate injection but 44.1% ± 11.7% 76 h after treatment with 188Re-labeled 4-hexadecyl-1,2,9,9-tetramethyl-4,7-diaza-1,10-decanethiol/lipiodol (188Re-HDD/lipiodol [iodized oil; Guerbet]) (2). However, it seems more relevant to compare the urinary elimination of 188Re in this context with the results obtained using 131I-lipiodol. If the shorter physical half-life of 188Re (16.9 h) is considered, our value of 44.1% over 76 h compares favorably with the observations of Raoul et al. and Nakajo et al. with 131I-lipiodol (3,4).
It was not specified how the 188Re-labeled human serum albumin (HSA) microspheres were administered. We applied an administration in the proper hepatic artery or both left and right branches, aiming at whole-liver treatment. Other authors have proposed injection as close to the tumor-feeding artery as possible (5). The administration protocol might have an impact on clearance of the radiopharmaceutical. Second, we assume that the presence of arteriovenous shunting in the liver affects elimination of the radionuclide after intraarterial administration. The degree of shunting depends on the underlying liver disease and, for patients with hepatocellular carcinoma and cirrhosis, is expected to be considerably higher than for patients with colorectal metastasis without underlying cirrhosis (6). Hence, the relatively low value of 8.5% of injected 188Re retrieved in the urine after administration of 188Re-labeled HSA microspheres in a mixed patient population (hepatocellular carcinoma and liver metastasis) is difficult to compare with the results obtained in our series, consisting of cirrhotic patients with hepatocellular carcinoma.
In addition, low urinary excretion of radionuclide does not guarantee good liver or tumor retention of the radiopharmaceutical. It would have been more relevant to compare the effective liver half-life rather than the effective half-life in the whole body. Was fecal elimination checked in the clinical study?
We agree that the low radiochemical yield of the 188Re-HDD/lipiodol labeling procedure is an important drawback to implementing high-activity treatment routinely. Probably, the kit presented by Wunderlich et al. (7) is also safer from a radioprotection point of view. In addition, the 188Re-labeled HSA microspheres have several interesting features such as a homogeneous particle size of 25 μm, an only-temporary occluding effect on the blood vessels, and a production cost that is probably lower than that for 90Y-labeled microspheres.