The Role of Amifostine as a Radioprotector

The article by Drs. Wasserman andBrizel concludes that protection of normal tissue from side effects due toradiotherapy will permit dose escalation and improve therapeutic efficacy andtreatment outcomes. Although this concept is sound in principle, its generalizedclinical application hinges critically on two controversial issues associatedwith the use of all putative radioprotective agents: (1) the extent of areduction in normal tissue side effects afforded in a variety of dose-limitingnormal tissues, and (2) the potential for tumor protection.

Modulation of Radiation Response

Although a great deal of effort has been expended on variousapproaches to sensitizing tumors to radiotherapy, attempts to protect normaltissue from ionizing radiation damage have focused primarily on sulfhydryl-containingcompounds.[1] A lead agent initially developed at the Walter Reed Army Instituteof Research (WR-2721) and subsequently known as amifostine (Ethyol) protectsagainst sparsely ionizing radiation predominantly by scavenging freeradicals.[1,2]

The extent of this radioprotection is strongly dependent onoxygen concentration, because the active metabolite of amifostine and oxygencompete for free radicals.[2] Indeed, in a series of elegant preclinicalstudies, Denekamp et al[3] clearly demonstrated that amifostine protection invivo is maximal at intermediate levels of oxygen. While the effective scavengingof free radicals offers an attractive model for the protective action of agentslike amifostine, this explanation is likely to be an oversimplification, withother complex factors undoubtedly being involved.[1,4]

Protection of Normal Tissues

Preclinical investigations have shown that, in general,amifostine offers good protection to the hematopoietic system and salivaryglands (protection factors near 3), intermediate protection to organs such asthe kidneys and lungs (protection factors 1-2), and no protection to thecentral nervous system.[1,4] In addition to this intertissue variability, theextent of amifostine protection within normal tissues varies. The nature of thevariability is uncertain, but factors such as differences in tissue oxygentensions as well as amifostine distribution and pharmacodynamics may beinvolved.[2,5]

As discussed by Drs. Wasserman and Brizel, the use ofamifostine may reduce the side effects of radiotherapy alone or in combinationwith chemotherapy in a number of clinical settings. In the largest study to date—aphase III randomized trial in head and neck cancer—Brizel et al[6] reportedthat patients who received amifostine had reductions in both acute xerostomia ofgrade 2 or greater (78% vs 51%, P < .0001) and chronic xerostomia ofgrade 2 or greater (57% vs 34%, P < .002). Median saliva production was alsogreater in patients treated with amifostine. However, no reduction inradiation-induced mucositis was seen.

Long-term xerostomia is associated with difficulty inmaintaining a normal diet and an increased risk of dental caries. Amifostine isclearly an option that may reduce the likelihood and severity of xerostomia and,thereby, improve quality of life. Another alternative is to reduce the treatmentvolume with conformal radiation therapy approaches. Intensity-modulatedradiation therapy and three-dimensional radiation therapy techniques may notonly reduce the incidence of acute and chronic xerostomia, but may also reducethe likelihood of other acute and late adverse events.

Therapeutic Index

The question of selectivity lies at the heart of the matterwhen discussing the use of radioprotectors in cancer treatment. In the early1980s, this issue sparked a firestorm of scientific debate.[7,8] Almost 20 yearslater, concerns that amifostine may protect tumors from the effects of radiationhave not been unequivocally alleviated. While some preclinical data—particularlythose obtained from trials using large single-drug doses—may be brought toquestion, results reported in studies in which such issues were not relevantcannot be so easily dismissed.[4] Indeed, the observation that more protectionmay be conferred on well-oxygenated tumors, as well as tumors treated withfractionated radiotherapy,[2,9,10] remains worrisome.

Clinical studies examining the impact of amifostine on thetherapeutic index are limited. In their article, Drs. Wasserman and Brizelconclude that there is no evidence in the literature that amifostine leads totumor protection. However, most of these published studies are phase I/II orretrospective. The number of patients in the treatment arms is usually modest(typically 20 to 50), follow-up times are often relatively short, and thetreatment administered may be quite variable in a given trial. All these factorsincrease the difficulty of evaluating these studies. Thus, while the availabledata suggest no apparent differences in tumor response rates, the statisticalpower of many of these studies is poor, and a 10% to 15% difference inlocoregional control could escape detection.[11]

Conclusions

According to Hall,[1]

. . . [O]ne of the difficult and worrisome factors in the use of radioprotectors in the clinic is that their use is not fail-safe. To exploit a benefit, radiation doses must be increased with the confidence that normal tissues are protected and that the extra dose can improve tumor response. If radioprotection does not occur, unacceptable normal tissue injury results.

These statements are particularly relevant in light of thevariable protection factors reported for amifostine in tissues treated withradiation. As noted by Lindegaard and Grau,[4] "A positive therapeuticindex obtained with regard to some normal tissues, like the salivary glands, maybe offset by a negative therapeutic index for the same treatment with regard toother normal tissues, like the spinal cord."

The findings of Brizel et al[6] clearly are encouraging andmay ultimately lead to new therapeutic options in patients who will experiencesignificant xerostomia after radiation therapy. Indeed, if subcutaneousinjections prove to be equally efficacious, this route of administration willsignificantly enhance the ability to administer this agent expeditiously. Still,given that amifostine has shown variable protection factors in an assortment oftissues treated with radiation, and in the absence of clinical studies withpatient numbers large enough to provide sufficient statistical power to detectmodest changes in tumor control, it would be prudent to refrain from using aradioprotective agent to allow irradiation dose escalation outside a controlledclinical trial.

References:

1. Hall EJ: Radiobiology for the Radiologist, 5th ed.Philadelphia, Lippincott Williams & Wilkins, 2000.

2. Travis EL: The oxygen dependence of protection byaminothiols: Implications for normal tissues and solid tumors. Int J RadiatOncol Biol Phys 10:1495-1501, 1984.

3. Denekamp J, Michael BD, Rojas A, et al: Radioprotection ofmouse skin by WR-2721: The critical influence of oxygen tension. Int J RadiatOncol Biol Phys 8:531-534, 1982.

4. Lindegaard JC, Grau C: Has the outlook improved foramifostine as a clinical radioprotector? Radiother Oncol 57:113-118, 2000.

5. Yuhas JM, Afzal SM, Afzal V: Variation in normal tissueresponsiveness to WR-2721. Int J Radiat Oncol Biol Phys 10:1537-1539, 1984.

6. Brizel DM, Wasserman TH, Henke M, et al: Phase IIIrandomized trial of amifostine as a radioprotector in head and neck cancer. JClin Oncol 18:3339-3345, 2000.

7. Denekamp J, Stewart FA, Rojas A: Is the outlook grey forWR-2721 as a clinical radioprotector? Int J Radiat Oncol Biol Phys 9:1247-1249,1983.

8. Yuhas JM: Efficacy testing of WR-2721 in Great Britain:Everything is black and white at the gray lab. Int J Radiat Oncol Biol Phys9:595-598, 1983.

9. Milas L, Hunter N, Ito H, et al: Effect of tumor type,size, and end point on tumor radioprotection by WR-2721. Int J Radiat Oncol BiolPhys 10:41-48, 1984.

10. Rojas A, Stewart FA, Soranson JA, et al: Fractionationstudies with WR-2721: Normal tissues and tumour. Radiother Oncol 6:51-60, 1986.

11. Mendenhall WM, Million RR: Elective neck irradiation forsquamous cell carcinoma of the head and neck: Analysis of time-dose factors andcauses of failure. Int J Radiat Oncol Biol Phys 12:741-746, 1986.