RHM

Ttwlttmioh

1at

D

A

1d

ationale, Conduct, and Outcome Usingypofractionated Radiotherapy in Prostate Cancer

ark Ritter, MD, PhD

Hypofractionated radiation therapy for prostate cancer has become of increasing interestwith the recognition of a potential improvement in therapeutic ratio with treatments deliv-ered in larger-sized fractions. In addition, the associated reduction in fraction numberproduces attractive cost and patient convenience advantages as well. A still limited butgrowing number of hypofractionation trials have reported acceptable short-term levels oftoxicity and biochemical control, but most have insufficient follow-up to ensure the long-term safety and efficacy of this approach. This situation will improve as many currentlyactive trials mature, particularly several high-value randomized trials. In contrast, extremehypofractionation, with schedules delivering only on the order of 5 fractions, is truly in itsinfancy for prostate cancer, with extremely limited tolerance and efficacy informationcurrently available. Several uncertainties in the radiobiology of hypofractionation mitigatefor an organized, cautious investigational approach. The fractionation response (�/� ratio)of prostate cancers and, for that matter, late-responding normal tissues, has yet to berigorously defined. Additionally, the linear-quadratic (LQ) model used in the design ofhypofractionation schedules is subject to its own uncertainties, particularly with respect tothe upper limit of fraction sizes for which it remains valid. Contemporary dose-escalatedradiation therapy is already highly effective, making it imperative that ongoing and futurestudies of hypofractionation be performed in carefully designed, randomized clinical trials.Clinical validation permitting, the adaptation of hypofractionation as a standard of carecould profoundly influence future management of localized prostate cancer.Semin Radiat Oncol 18:249-256 © 2008 Elsevier Inc. All rights reserved.

tstt

ccovppbiaf

fpvo

here now is convincing evidence that biochemical con-trol is improved with higher cumulative doses of radia-

ion to the prostate. This dose escalation can be accomplishedithout undue complications by improving the physical de-

ivery of the radiation using 3-dimensional conformal radio-herapy or, better, intensity-modulated radiation therapyechniques. However, an unusual prostate tumor radiobiologyay allow a radically different approach to dose escalation that

s radiobiological in nature. The unusual aspect of this radiobi-logy relates to prostate cancer’s suspected, uncharacteristicallyigh sensitivity to large fractions of radiation.Conventional fractionation schemes using fraction sizes of

.8 to 2.0 Gy are based on the premise that tumors typicallyre less responsive to fraction size (have higher �/� ratios)han are late-responding normal tissues (with lower �/� ra-

epartment of Human Oncology and Medical Physics, University of Wis-consin School of Medicine and Public Health, Madison, WI.

ddress reprint requests to Mark Ritter, MD, PhD, Department of HumanOncology and Medical Physics, University of Wisconsin School of Med-icine and Public Health, 600 Highland Avenue, K4/B100, Madison, WI

c53792. E-mail: ritter@humonc.wisc.edu

053-4296/08/$-see front matter © 2008 Elsevier Inc. All rights reserved.oi:10.1016/j.semradonc.2008.04.007

ios). An improved therapeutic ratio is therefore usuallyought by using multiple, relatively small radiation fractionso leverage the lower �/� ratios that late-responding normalissues have relative to most classes of tumors.

In contrast, recent analyses and reviews of clinical tumorontrol data have argued for a low �/� ratio for prostateancer on the order of 1 to 3 Gy, not the typical ratio of 8 Gyr higher that may apply to other tumors. This is also a loweralue than is typically ascribed to the adjacent organs at risk,rimarily bladder and rectum, putatively because the slowroliferation rate characteristic of most prostate tumors maye conducive to repair of radiation damage. This characteristic,

f true, would imply a unique opportunity to improve the ther-peutic ratio by treating prostate cancers with fewer but largerractions of radiation, a hypofractionation approach.

This article includes (1) a brief review of the evidenceavoring or challenging an altered fractionation response forrostate cancer; (2) a description of the predicted impact ofarious hypofractionation approaches on tumor control andn normal tissue late toxicities; (3) a review of ongoing and

ompleted clinical trials that use hypofractionation; and (4) a

249

dtr

dhte

TWRC1anrnfrwtpt8une

stasgTstdissf

dBfapfmtTcaP

c

dor

vpciffpsmloauwarpadtp

ahepwlawep

CTSlsouaptrtsmlia�

p

250 M. Ritter

escription of ongoing treatment regimens that are extendinghe concept to extreme hypofractionation, a stereotactic bodyadiotherapy approach.

The opportunity will also be taken to emphasize that ra-iobiological uncertainties remain in the models applied toypofractionation, uncertainties that necessitate caution andhe need for well-defined clinical trials, particularly whenxtreme hypofractionation is being used.

he Case for Hypofractionationhat Is the Fractionation

esponse of Prostate Cancer?onventional fractionation schemes using fraction sizes of.8 to 2.0 Gy are based on the premise that tumors typicallyre less responsive to fraction size than are late-respondingormal tissues. The �/� ratio is a measure of fractionationesponse, with low ratios associated with late-respondingormal tissues. A low �/� is consistent with a greater capacity

or repair between fractions, with an accompanying greaterelative sparing with small fraction sizes, than for tumorsith their typically higher �/� ratios. Under these condi-

ions, an improved therapeutic ratio is achieved with multi-le small fractions for most types of tumors. The �/� ratioshought to be associated with tumors, however, are typicallyor greater, whereas for late responding normal tissues, val-es on the order of 3 or 4 or somewhat less for the centralervous system are suggested from the analyses of numerousxperimental and some clinical outcome studies.

There appear to be exceptions to such typical tumor re-ponse to fractionation, however. Growth fraction (or effec-ive cell-cycle time) has often been associated with fraction-tion response, with slowly proliferating normal tissues (andome slowly proliferating tumors) generally displaying stron-er than expected fraction size responses (low �/� ratios).his relationship has been shown for melanomas1 and forome sarcomas,2 for example. In the case of prostate cancer,here is ample evidence for slow proliferation, based both onirect measurement of potential doubling times and labeling

ndices3 and on analysis of the kinetics of rising prostatepecific antigen (PSA) during tumor recurrence.4 Whetheruch is the case in all prostate cancers, those with high grade,or example, is not clear.

Recent analyses and reviews of clinical tumor responseata do in fact argue for a low �/� ratio for prostate cancer.5,6

renner and Hall,5 for example, analyzed dose response dataor external beam radiation compared with I-125 brachyther-py data and estimated a very low �/� ratio of 1.5 Gy forrostate cancer. Duschene and Peters6 argued in analogousashion that the �/� ratio for prostate cancer may be low and

ore similar to that expected for late-responding normalissue than for the typical, more rapidly proliferating tumor.hese studies involved very simple calculational approachesomparing 145 Gy given at a low dose rate with 70 to 74 Gyt fractionated high dose rates for patients with similar initialSA levels and Gleason scores.Fowler et al7 also conducted a comprehensive analysis of

linical outcome in patients treated with external-beam ra- b

iotherapy only, I-125 implants only, or Pd-103 implantsnly to further test the previously described analyses. An �/�atio for prostate cancer well below 2 Gy was also estimated.

Perhaps more convincing evidence for a low �/� is pro-ided in the data of Brenner et al.8 In this study, patients withrostate cancer were treated with a standard external-beamourse of treatment followed by high-dose rate temporarymplant boost doses, which were escalated by decreasingraction number from 3 to 2 and by increasing fraction sizerom 5.5 to 10.5 Gy. Patients were grouped according torognostic factors and biochemical control was modeled ver-us equivalent dose, as calculated via a linear-quadraticodel. Higher biochemical control rates observed with esca-

ation of hypofractionation were consistent with an �/� ratiof 1.2 (95% confidence interval, 0.03-4.1 Gy), again very lownd in approximate agreement with values reported earliersing other estimation approaches. This study’s advantageas its ability to compare several high-dose rate brachyther-

py regimens that differed only in the radiation fraction size,ather than needing to compare 2 diverse treatments such asrostate seed brachytherapy and external-beam radiother-py. Its limitation was the uncertainty of accounting for theose heterogeneity that is intrinsic to high-dose rate brachy-herapy, a limitation that is present in the low-dose rate im-lant analyses previously described as well.Another study using both external-beam and brachyther-

py data is that of Williams et al,9 who used a proportionalazards model to estimate the �/� ratio from data on 3,756xternal-beam and 185 high-dose rate brachytherapy boostatients. An estimated ratio of 2.6 Gy was determined butith a wide 95% confidence interval of 0.9 to 4.8 Gy. The

imited range of external-beam fraction sizes as well as patientnd prescription dose heterogeneity restricted the precisionith which the �/� ratio could be estimated. Other efforts to

stimate the �/� ratio from published clinical outcomes ofurely external-beam studies are described in detail later.

hallenges to the Concepthat �/� for Prostate Cancer Is Loweveral authors have expressed concern that potential prob-ems with analyses like those mentioned earlier could lead toubstantial uncertainties in the values derived for �/�. Somef these concerns, as previously described, relate to potentialncertainties with respect to the variable dose distributionsnd dose rates that occur within implants, factors not ex-ected to play as significant a role with external-beam radio-herapy, as well as in potential differences in the effectiveadiation quality of implants versus external-beam radia-ion.10 Altough these are certainly issues that warrant discus-ion, it is also true that when clinical parameters used in theodeling are restricted to reasonable ranges, an �/� ratio of

ess than 3 usually results. In fact, adjusting for these factorsn the most straightforward fashion tends in some cases toctually reduce rather than increase the calculated value of/� for prostate tumors.In addition, the potential impact of tumor clonogen re-

opulation during the protracted delivery of permanent seed

rachytherapy has been raised by Wang et al11 as a potential

cbcwfqg2sctow

becvhwefmuf

ctbrroa

TftAftrtflbtar

Apaa

tcnidp

ntfe

twttaacaagde

shilrbfa

Fmrar

Hypofractionated radiotherapy in prostate cancer 251

onfounding factor in the derivation of the �/� ratio fromrachytherapy data. In their analysis, if repopulation oc-urred during treatment, the resultant calculation for �/�ould yield a value of about 3 instead of the value of 1.5

ound otherwise. Although a thoughtful analysis, a potentialuestion arises from the authors’ use of relatively short clono-en repopulation initiation times of 0 or 28 days. Although8 days may be appropriate for a rapidly proliferating tumoruch as in the head and neck, it may be too short for prostateancer, given this tumor’s long average potential doublingime of about 35 days.3 A repopulation kickoff time in excessf 100 days appears more likely, at which point repopulationould no longer significantly influence an �/� calculation.Another challenge to the estimate of a low �/� ratio has

een raised by Nahum et al,12 who hypothesize that the pres-nce of hypoxia-related radioresistance in prostate cancerould produce significant underestimates of the �/� ratio,alues of 3 or less, for example, instead of actual values of 8 origher, if hypoxia is not taken into account. Several potentialeaknesses in this hypothesis have been discussed,13 how-

ver, including its reliance on survival parameters derivedrom in vitro prostate cancer cell-survival curves. In addition,

odeling tumor control probabilities with hypoxia can benreliable, particularly given the potentially confoundingactors of reoxygenation or inhomogeneity.13

With all information, both pro and con, taken into ac-ount, a low �/� ratio for prostate cancer remains an attrac-ive hypothesis that is supported by several lines of evidence,ut uncertainties clearly remain that will ultimately not beesolved until biochemical control data from large, preferablyandomized hypofractionation studies with 5 or more yearsf follow-up become available. As presented later, such datare now beginning to emerge.

he Theoretical Potentialor Hypofractionationo Improve Tumor Controllthough conventional fractionation schemes using small

raction sizes of 1.8 to 2.0 Gy are based on the premise thatumors typically have high �/� ratios that make them lessesponsive to fraction size than are late-responding normalissues, the situation may well be reversed for prostate cancer,avoring the use of hypofractionation. These relationships, ateast over a range of fraction sizes between 1 and 6 Gy, canest be shown through the use of the linear-quadratic equa-ion, which calculates the biologically effective dose, BED, forgiven total dose, D, dose per fraction, d, and alpha-beta

atio, �/�:

BED � D�1 �d

� ⁄ ��n �/� for tumor less than that for at-risk normal tissuesredicts an improved therapeutic ratio with hypofraction-tion. For example, if the ratio of biologically effective dose at

n �/� of 1.5 for tumor versus 3 for late tissue toxicity is i

aken as a form of therapeutic ratio, then this ratio, readilyalculated using the linear-quadratic equation, increases sig-ificantly with fraction size (Fig 1). Although the relationship

s mathematically independent of the total dose, the totalose would of course need to be appropriately limited torevent undue toxicities.If, on the other hand, �/� ratios for prostate cancer and late

ormal tissue damage were equal, there then would be nei-her gain nor loss in predicted therapeutic gain from hypo-ractionation, although improvements in convenience andfficiency could still be useful benefits.

Some controversy exists regarding the upper range of frac-ion sizes for which the linear-quadratic model remains valid,ith some support for it being applicable to very large frac-

ion sizes, even 20 to 30 Gy,14,15 particularly with modifica-ions as needed to account for processes such as reoxygen-tion and redistribution,16 but other analyses suggesting that,lthough the linear-quadratic model is likely to be suffi-iently accurate for fraction size ranges up to 6 or 7 Gy,pplication of the model to larger fraction sizes could lead ton underprediction of the total dose required to produce aiven effect.17 Such an underprediction of total requiredose, if acted on, could lead to a less toxic but also lessffective treatment.

Despite these remaining uncertainties, however, there isufficient supporting evidence to justify continuing to test theypothesis that larger radiation fraction sizes will selectively

ncrease prostate tumor-cell kill relative to the induction ofate effects, offering the promise of an improved therapeuticatio with hypofractionation. Although many were plannedefore an advantageous response of prostate cancer to largeractions was suspected, a number of prostate hypofraction-tion trials have been performed and information regarding

igure 1 The ratio of the biologically effective dose for prostate tu-or (�/� � 1.5) to normal tissue late effects (�/� � 3). This pa-

ameter could be regarded as a crude estimator of therapeutic rationd is shown as a function of fraction size. The maximum theoreticalatio equals the ratio of the normal tissue to tumor �/� ratio, which

n this case is 2. (Color version of figure is available online.)

tc

HHETtdwtms

AoTtngultbpl

f

e(avsclao

AEAtetd

s12tcbfviifes

Frtt2Gxr

Ftic1deh

252 M. Ritter

he efficacy and safety of such an approach is emerging. Suchompleted and ongoing efforts will be detailed later.

ow Best Canypofractionation Bexplored in a Clinical Setting?

wo types of hypofractionation designs could be consideredhat would exploit the hypothesized radiobiological advantagesescribed earlier: (1) normal tissue de-escalation of total dosehile maintaining constant predicted tumor control and (2)

umor biological dose escalation with constant predicted nor-al tissue late effects. These 2 hypofractionation approaches

eek to achieve different desirable objectives.

pproach 1: Normal Tissue De-escalationf the Total Dose With Constant Tumor Controlhis approach starts with the premise that a certain well-

ested 1.8- to 2-Gy per fraction scheme and treatment tech-ique provides an acceptably high tumor control rate for aiven group of patients but, at the same time, produces anndesirable level of late complications, say grade II or higher

ate rectal bleeding. Assuming a tumor �/� of 1.5 and a lateissue �/� of 3, a schema of dose per fraction escalation cane proposed, with linear-quadratic modeling, that wouldredict a reduction in side effects while maintaining the same

evel of tumor control (Fig 2).For example, a hypofractionation step from 76 Gy in 38

igure 2 An example of a dose per fraction escalation schemes thateduce the effective normal tissue dose for late effects while main-aining a constant predicted level of tumor control. The late NTD ishe cumulative dose adjusted to its equivalent dose if delivered in-Gy fractions to late-responding normal tissues. A tumor �/� of 1.5y and a late tissue toxicity �/� of 3 Gy are assumed. The upper-axis shows the actual cumulative dose delivered by each of theegimens.

ractions of 2 Gy each to 58.8 Gy in 20 fractions of 2.94 Gy fi

ach would predict a late normalized total dose (NTD) doseequivalent to a dose delivered in 2-Gy fractions) of onlybout 70 Gy without loss of tumor control. A potential ad-antage of such an approach would be that the late toxicityparing achieved could allow use of a less sophisticated, lessonformal treatment technique while attaining the same highevel of uncomplicated tumor control. The actual outcome oftrial based on this approach would, of course, be dependentn the accuracy of the assumed �/� ratios.

pproach 2: Tumor Biological Dosescalation With Constant Late Effectspotentially more interesting approach is to devise a frac-

ion-size escalation schedule that, as calculated using the lin-ar-quadratic equation, would predict a gain in tumor con-rol while maintaining a constant, biologically effective Gy3

ose for late-responding normal tissues (Fig 3).Starting with an already dose-escalated fractionation

cheme of 38 � 2 Gy fractions and assuming a tumor �/� of.5, it is seen that the equivalent total dose normalized to-Gy fractions (NTD3) increases substantially with hypofrac-ionation (x-axis) even as the actual total delivered dose de-reases. The corresponding predicted tumor control proba-ility, derived from Fowler et al18, also increases. An �/� ratioor prostate cancer lower than that for normal tissues pro-ides the basis for improving tumor control without increas-ng late-effect risk. If tumor and normal tissue �/�’s werenstead equal, tumor control would not improve with hypo-ractionation, but the cost and convenience benefit of deliv-ring fewer fractions would remain. Of note, the 28 � 2.5 Gychedule in Figure 3 has been used for some time by Cleve-

igure 3 Dose per fraction escalation regimens predicted to increaseumor control while maintaining constant late normal tissue toxic-ty. Biochemical freedom from disease (bNED) is estimated by cal-ulating the equivalent doses if delivered in 2-Gy fractions (�/� �.5) and determining the corresponding biochemical freedom fromisease values from the biochemical control versus dose data in Fowlert al.18 The total dose delivered is seen to decrease with increasingypofractionation to maintain constant late effects. (Color version of

gure is available online.)

loOe

WSWfHbUatntTppvasc

fifastf

cCCblcCacaa

w�aME

T

5

6

6

5

5

6

7

655

565677

7

7

56

Hypofractionated radiotherapy in prostate cancer 253

and Clinic investigators and has been extensively reportedn.19,20 The recently opened, randomized Radiation Therapyncology Group trial 0415 has adopted this regimen in its

xperimental, hypofractionation arm.

hat Is the Currenttate of Clinical Experienceith Hypofractionation

or Prostate Cancer?ypofractionated external-beam radiotherapy has actuallyeen used clinically for a number of years, particularly in thenited Kingdom.21-23 Although these treatments were gener-lly well tolerated, overall efficacy is difficult to assess, givenhat these trials were performed largely in the pre-PSA era. Aumber of more contemporary hypofractionation trials, ei-her published or currently in progress, are shown in Table 1.he most straightforward and intuitive way of estimating theredicted effectiveness and toxicities of these various ap-roaches is to use the linear-quadratic modeling to equate thearious hypofractionation practices to the normalized equiv-lent dose (NED) if delivered in 2-Gy fractions. These arehown in Table 1 for assumed �/� of 1.5 and 3 for prostate

able 1 Hypofractionation Trials: Schedules and Equivalent T

Fractionation(tot.dose/fx size/

#fx)

Total Dose Equivalent if Givein 2-Gy Fractions (NTD2)

Alpha/Beta �1.5 (tumor)

Alpha/Beta �3 (late

complications

0 Gy/3.13 Gy/16 fx 66 Gy 61.3 Gy

9 Gy/3 Gy/23 fx 88.7 Gy 82.8 Gy

6 GyE/3.3 GyE/20 fx(carbon ions)

90.5 Gy 83.1 Gy

2.5 Gy/2.625 Gy/20 fx 61.9 Gy 59.1 Gy

6 Gy/3.5 Gy/16 80 Gy 72.8 Gy

0 Gy/3 Gy/20 fx 77.2 Gy 72 Gy

0 Gy/2.5 Gy/28 fx 80 Gy 77 Gy

4.7 Gy/2.94 Gy/22 fx 82.6 Gy 77 Gy8.1 Gy/3.63 Gy/16 fx 85.1 Gy 77 Gy1.6 Gy/4.3 Gy/12 fx 85.5 Gy 75 Gy

2.5/2.625 Gy/20 fx 61.9 Gy 59.1 Gy6 Gy/2 Gy/33 fx 66 Gy 66 Gy5 Gy/2.75 Gy/20 fx 66.8 Gy 63.2 Gy4 Gy/2 Gy/32 fx 64 Gy 64 Gy0.2 Gy/2.7 Gy/26 fx 84.2 Gy 80 Gy6 Gy/2 Gy/38 fx 76 Gy 76 Gy

0 Gy/2.5 Gy/28 fx 80 Gy 77 Gy

3.8 Gy/1.8 Gy/41 fx 69.6 Gy 70.8 Gy

7 Gy/3 Gy/19 fx 73.3 Gy 68.4 Gy0 Gy/3 Gy/20 fx 77.2 Gy 72 Gy

ancer and late-responding normal tissue, respectively. It is b

rst of all apparent that, even for these only modestly hypo-ractionated schedules, normalized doses range betweenbout 4% and 8% higher for tumor than for normal tissue,howing the potential for therapeutic gain even with rela-ively modest hypofractionation should prostate cancer inact have a lower �/� ratio than normal tissue.

Of the previously described trials that are contemporary, areompleted, and are reported, only the Princess Margaret,29

leveland Clinic,20 Manchester,24 National Cancer Institute-anada,31 and Chiba carbon ion26 trials have sufficient num-ers of patients and sufficient albeit still relatively short fol-

ow-up to enable preliminary estimates of biochemicalontrol and toxicity. Of these 5, only the Princess Margaret,leveland Clinic, and Chiba carbon ion trials deliver equiv-lent doses sufficiently large enough (assuming �/� � 1.5) tourrently be considered adequately dose escalated. Addition-lly, only these 3 trials included sufficient patient numbersnd follow-up to adequately estimate late toxicity.

Reported toxicity in these 3 trials was quite acceptable,ith the actuarial Radiation Therapy Oncology Group grade2 late rectal and genitourinary toxicity rates being only 2%

nd 2%, 4.5% and 5.3%, and 1% and 1% for the Princessargaret, Cleveland Clinic, and Chiba trials, respectively.fficacy also has appeared to be satisfactory. An analysis of

oses in 2-Gy Fractions

o. of PTS Institution References

705 Christie Hosp,Manchester

Livsey et al24

52 Gunma, Japan Akimoto et al25

201 NIRS, Chiba,Japan

Tsuji et al26

300 Edenburgh Higgins et al27

36 Jette, Belgium Soete et al28

92 Princess Margaret Martin et al29

770 Cleveland Clinic Kupelian et al19,20

100 Multi-institutionaltrial

Ritter et al30

100(accruing)

466 NCI-Canada(phase III)

Lukka et al31

470108 Adelaid (phase III) Yeoh et al32

109150 Fox Chase

(phase III)Pollack et al33

150

ngoing (goalof 1,067 pts)

RTOG 0415(phase III)

www.rtog.org/members/protocols/0415/0415.pdf

ngoing (goalof 2,100 pts)

MRC (phase III) Khoo et al34

otal D

n

N)

58

O

O

iochemical control in the Cleveland Clinic trial yielded a

5v2c

ear2ucidacd

epFdTtraivcr1

tddaic

oo1

iOcIrstettaiosa

c(rtatiictmc

plrfsactaqitms

EE

Fdipfrfdap3a

254 M. Ritter

-year overall rate of 82%, equivalent to or better than pre-iously attained with the institutional standard of 78 Gy in-Gy fractions.20 The Chiba trial found an overall biochemi-al control rate of 83% at 5 years.

The Manchester study24 used a schedule delivering lowerquivalent doses and accordingly reported low late toxicitiesnd relatively poor biochemical control rates, consistent withesults expected for a “conventional” doses of only 66 Gy in-Gy fractions. Similarly, the NCI-Canada trial used a sched-le equivalent to only about 62 Gy in 2-Gy fractions.31 Bio-hemical control rates and toxicities were also correspond-ngly low. Despite the low equivalent doses these 2 trialselivered, comparisons of their reported outcomes can en-ble a useful testing of hypofractionation modeling. The Prin-ess Margaret and Cleveland Clinic studies, with their higher-elivered equivalent doses, are valuable for such testing as well.A graphic depiction of biochemical control rates versus

quivalent dose from these 4 studies (for intermediate-riskrostate cancer, when specifically reported) is presented inigure 4. Represented are 3- to 5-year actuarial biochemicalisease-free survival rates using the American Society forherapeutic Radiation Oncology definition, the only defini-

ion uniformly available in these reports. The solid line dose-esponse curve for radiation delivered in 2-Gy fractions isdapted from Fowler et al18 and is based on 5-year biochem-cal control data for intermediate-risk patients from 5 con-entionally fractionated prostate cancer trials. Biochemicalontrol points from the hypofractionation trials are plottedelative to their equivalent dose for 3 assumed �/� ratios of.5, 3, and 10. The ratio of 1.5 clearly produces the closest fit

igure 4 Biochemical disease-free survival rates versus equivalentoses from four hypofractionation studies identified in Table 1 (for

ntermediate-risk prostate cancer when separately reported in theublications). Shown are 3- to 5-year actuarial biochemical disease-

ree survival rates using the ASTRO definition. The solid line dose-esponse curve for radiation delivered in 2-Gy fractions is adaptedrom Fowler et al18 and is based on 3- to 5-year biochemical controlata for intermediate risk patients from 5 conventionally fraction-ted prostate cancer trials. Biochemical control points from the hy-ofractionation trials are plotted relative to their equivalent dose forassumed �/� ratios of 1.5, 3, and 10. (Color version of figure is

pvailable online.)

o the curve. Although many unaccounted for variables ren-er this a nonrigorous and strictly post hoc comparison, theegree of outcomes agreement between hypofractionatednd conventional regimens when an �/� of 1.5 is chosenntriguingly suggests that prostate cancer response is indeedharacterized by a low �/� ratio.

A more formalized analysis by Bentzen and Ritter35 of onef these trials, the NCI-Canada study,31 again but more rig-rously yields a quite low �/� ratio estimate for prostate of.12 Gy with 95% confidence interval (�3.3 to 5.6 Gy).The last nonrandomized study listed in Table 1 is a multi-

nstitutional trial (University of Wisconsin, MD Anderson-rlando, Wayne State University, Medical College of Wis-

onsin, and JT Vucurevich Cancer Institute) and is a phase/II study30 that escalates dose per fraction in 3 steps, with lateectal bleeding being the escalation-limiting factor. The de-ign results in predicted late effects expected to remain rela-ively constant (at a level consistent with about 76 Gy deliv-red in 2-Gy fractions) even as fraction size escalates. Therial design also includes a nested fractions-per-week escala-ion/de-escalation to monitor for and prevent unacceptablecute toxicities that might result from too extreme a shorten-ng of treatment duration. Overly severe acute toxicities notnly create intratreatment morbidity but might also lead too-called consequential late injuries in adjacent organs suchs the rectum or bladder.18

Preliminary results from this phase I/II trial30 have indi-ated acceptably low rates of gastrointestinal and genitourinaryGU) toxicity (2-year grade 2 gastrointestinal and GU toxicityates of 8.8% and 3%, respectively, even after 5-day-per-weekreatment and, also, preliminary biochemical control rates thatre high and in the expected range. The trial is nearing comple-ion with 258 of a target 300 patients accrued. When mature,ts 3 increasingly hypofractionated schedules will be evaluablen terms of their suitability for future trials and, in addition,ollected and centrally analyzed dose-volume data together withhe trial’s 3 significantly different doses per fraction should per-it solid estimates to be made of �/� ratios both for prostate

ancer as well as for adjacent organs at risk.Thus, the outcomes of several hypofractionation trials sup-

ort the hypothesis that the �/� ratio for prostate cancer isow and that further investigation via prospective, preferablyandomized clinical trials is fully warranted. As is also evidentrom Table 1, several such randomized trials that deliveruitably high effective doses have either recently completedccrual or are currently underway. Thus, the prospects for aomprehensive evaluation of this potentially clinically advan-ageous, cost-effective, and convenient treatment approachppear excellent. Ideally, these or future trials will also ade-uately collect delivered rather than planned dose-volume

nformation, enabling the most accurate analysis of the frac-ionation response of prostate cancer and normal tissue. Theethodology to do so has become available, and its use

hould be encouraged.

xtreme Hypofractionationven fewer fractions are beginning to be used in some clinical

ractices, although not always in the context of prospective

to5timrmifAprasrvtct

titdfacap

htVfiarroa

Uifemtsmlsp

CRsidolhotiq

psptlif

R

T

3

3

4

3

455

* TDs fo

Hypofractionated radiotherapy in prostate cancer 255

rials. Called stereotactic body radiotherapy (SBRT) whennly 5 total fractions are used, fraction sizes of between about.5 and 7 Gy are typically given, although higher-dose frac-ions have been used as well. Although this further reductionn fraction number might seem a logical extension of current,

ore modestly hypofractionated studies, there are numerouseasons why such an approach is even more in need of theonitoring and reporting guidelines that are integral to clin-

cal trials. First, definitive results using more modest hypo-ractionation, although encouraging, are not yet mature.lso, unlike SBRT as used in treating lung cancer, for exam-le, the intent of prostate SBRT is not ablative, a restrictionelated to the urethra’s intraprostatic location. More moder-te but still large fraction sizes in the 5-to 7-Gy range oromewhat higher are therefore considered, but uncertaintiesemain over the validity of the linear-quadratic model orariations thereof for predicting biological effectiveness ofhese higher fraction sizes. Radiation damage mechanismsould change with increasing fraction size, rendering predic-ions from such models unreliable.

Furthermore, the potential tumor control-enhancing con-ributions of reoxygenation and redistribution could dimin-sh as the number of fractions decreases and the duration ofreatment shortens. In addition, the demand for treatmentelivery accuracy surely intensifies with the delivery of fewer

ractions, making it imperative that quality-control issues suchs immobilization, target motion, and image guidance be givenonsiderable attention. Only prospective trials can adequatelyddress these requirements and ensure that patient safety androper documentation of outcomes are provided for.There are several studies of extreme hypofractionation,

owever, that are being performed appropriately in prospec-ive fashion. Four are listed in Table 2, of which only theirginia Mason trial has been completed36 and has sufficient

ollow-up (median 48 months) to allow meaningful report-ng. Reported outcomes were acceptable: Actuarial late GUnd GI grade 2 toxicities at 48 months were 16.1% and 9.4%,espectively, whereas the actuarial freedom from biochemicalelapse at 48 months was 90% (nadir plus 2). The Universityf Toronto phase I/II trial has only reported on 30 patients

able 2 Phase I/II Ultrahypofractionation Trials: Schedules an

Fractionation(tot.dose/fx size/#fx)

Total Dose Equivalent in 2Gy Fractions (NTD2)

Alpha/Beta �1.5 (tumor)

Alpha/Beta �3 (late effects

3.5 Gy/6.7 Gy/5 fx 78 Gy 64.9 Gy

6.25 Gy/7.25 Gy/5 fx 90.6 Gy 74.3 Gy

2.7 Gy/6.1 Gy/7 fx 92.7 Gy 77.7 Gy

5 Gy/7 Gy/5 fx 85.1 Gy 70 Gy

7.5 Gy/9.5 Gy/5 fx 149 Gy* 118 Gy0 Gy/10 Gy/5 fx 164 Gy 130 Gy2.5 Gy/10.5 Gy/5 fx 180 Gy 142 Gy

NTD doses based on linear-quadratic modeling may overpredict N

nd only on acute toxicities, which were acceptably low. The

niversity of Texas Southwestern Medical Center study listedn Table 2 is a recently initiated, ongoing phase I/II study notableor the higher-dose fractions it uses, specifically 5 fractions ofither 9.0, 9.5, or 10 Gy each. Were standard linear-quadraticodeling to remain valid at these much larger fraction sizes,

hese regimens could produce very high tumor and normal tis-ue equivalent doses of 135 and 108 Gy, respectively, althoughodified linear-quadratic models17 would predict significantly

ower equivalent doses. Furthermore, normal tissue volume re-trictions in place during treatment planning could reduce theotential for toxicity from these high dose fractions as well.

onclusionsesults from several mostly phase I/II prostate radiotherapytudies have indicated a gain in therapeutic ratio with increasesn radiation fraction size beyond the 1.8 to 2 Gy typical of stan-ard practice. However, given that uncertainties exist in extrap-lating biological effects to larger fractions and that dose-esca-ated radiation therapy with standard fractionation is alreadyighly effective, it is imperative that ongoing and future studiesf hypofractionation be performed in a randomized controlledrial setting. Ideally, delivered rather than planned dose-volumenformation would be recorded in such studies to maximize theuality of fractionation response information obtained.Several large, randomized trials are underway or are com-

leted and awaiting sufficient follow-up for analysis, and it istudies such as these that will ultimately define the utility ofrostate hypofractionation. Hypofractionation has the poten-ial for significant therapeutic gain as well as economic andogistic advantage. Should the approach be clinically validated,ts acceptance as a standard of care will profoundly influence theuture management of localized prostate cancer.

eferences1. Bentzen SM, Overgaard J, Thames HD, et al: Clinical radiobiology of

malignant melanoma. Radiother Oncol 16:169-182, 19892. Thames HD, Suit HD: Tumor radioresponsiveness versus fractionation

sensitivity. Int J Radiat Oncol Biol Phys 12:687-691, 19863. HaustermansKM,HoflandI,VanPoppelH,etal:Cellkineticmeasurements in

prostate cancer. Int J Radiat Oncol Biol Phys 37:1067-1070, 1997

ivalent Total Doses in 2-Gy Fractions

. of PTS Institution References

40 Virginia Mason Madsen et al36

(ongoing) Stanford Pawlicki et al37

105 Umea Widmark (personalcommunication, 2008)

(ongoing) University of Toronto Tang et al38

15 UTSW, Dallas Timmerman (personalcommunication, 2008)(ongoing)

—

r large fractions, as in the UTSW trial.

d Equ

No)

23

30

10

4. Pollack A, Zagars GK, Kavadi VS: Prostate specific antigen doubling

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3

3

3

256 M. Ritter

time and disease relapse after radiotherapy for prostate cancer. Cancer74:670-678, 1994

5. Brenner DJ, Hall EJ: Fractionation and protraction for radiotherapy ofprostate carcinoma. Int J Radiat Oncol Biol Phys 43:1095-1101, 1999

6. Duchesne GM, Peters LJ: What is the alpha/beta ratio for prostate can-cer? Rationale for hypofractionated high-dose-rate brachytherapy [ed-itorial]. Int J Radiat Oncol Biol Phys 44:747-748, 1999

7. Fowler J, Chappell R, Ritter M: Is alpha/beta for prostate tumors reallylow? Int J Radiat Oncol Biol Phys 50:1021-1031, 2001

8. Brenner DJ, Martinez AA, Edmundson GK, et al: Direct evidence thatprostate tumors show high sensitivity to fractionation (low alpha/betaratio), similar to late-responding normal tissue. Int J Radiat Oncol BiolPhys 52:6-13, 2002

9. Williams SG, Taylor JM, Liu N, et al: Use of individual fraction size datafrom 3756 patients to directly determine the alpha/beta ratio of prostatecancer. Int J Radiat Oncol Biol Phys 68:24-33, 2007

0. D’Souza WD, Thames HD: Is the a/b ratio for prostate cancer low? Int JRadiat Biol Oncol Phys 51:1-3, 2001

1. Wang JZ, Guerrero M, Li XA: How low is the alpha/beta ratio forprostate cancer? Int J Radiat Oncol Biol Phys 55:194-203, 2003

2. Nahum AE, Movsas B, Horwitz EM, et al: Incorporating clinical mea-surements of hypoxia into tumor local control modeling of prostatecancer: Implications for the alpha/beta ratio. Int J Radiat Oncol BiolPhys 57:391-401, 2003

3. Fowler JF, Nahum AE, Orton CG: Point/counterpoint. The best radio-therapy for the treatment of prostate cancer involves hypofractionation.Med Phys 33:3081-3084, 2006

4. Douglas BG, Fowler JF: The effect of multiple small doses of x rays onskin reactions in the mouse and a basic interpretation. Radiat Res 66:401-426, 1976

5. Hall EJ, Brenner DJ: Extrapolating hypofractionated radiation schemesfrom radiosurgery data: Regarding Hall et al. IJROBP 21:819-824; 1991and Hall and Brenner, IJROBP 25:381-385; 1993. Int J Radiat OncolBiol Phys 32:275-276, 1995

6. Brenner DJ, Hlatky LR, Hahnfeldt PJ, et al: A convenient extension ofthe linear-quadratic model to include redistribution and reoxygen-ation. Int J Radiat Oncol Biol Phys 32:379-390, 1995

7. Guerrero M, Li XA: Extending the linear-quadratic model for largefraction doses pertinent to stereotactic radiotherapy. Phys Med Biol49:4825-4835, 2004

8. Fowler JF, Ritter MA, Chappell RJ, et al: What hypofractionated proto-cols should be tested for prostate cancer? Int J Radiat Oncol Biol Phys56:1093-1104, 2003

9. Kupelian PA, Reddy CA, Carlson TP, et al: Preliminary observations onbiochemical relapse-free survival rates after short-course intensity-modulated radiotherapy (70 Gy at 2.5 Gy/fraction) for localized pros-tate cancer. Int J Radiat Oncol Biol Phys 53:904-912, 2002

0. Kupelian PA, Willoughby TR, Reddy CA, et al: Hypofractionated inten-sity-modulated radiotherapy (70 Gy at 2.5 Gy per fraction) for localizedprostate cancer: Cleveland Clinic experience. Int J Radiat Oncol BiolPhys 68:1424-1430, 2007

1. Collins CD, Lloyd-Davies RW, Swan AV: Radical external beam radio-therapy for localised carcinoma of the prostate using a hypofraction-

ation technique. Clin Oncol (R Coll Radiol) 3:127-132, 1991

2. Duncan W, Warde P, Catton CN, et al: Carcinoma of the prostate:Results of radical radiotherapy (1970-1985). Int J Radiat Oncol BiolPhys 26:203-210, 1993

3. Read G, Pointon RC: Retrospective study of radiotherapy in early car-cinoma of the prostate. Br J Urol 63:191-195, 1989

4. Livsey JE, Cowan RA, Wylie JP, et al: Hypofractionated conformalradiotherapy in carcinoma of the prostate: Five-year outcome analysis.Int J Radiat Oncol Biol Phys 57:1254-1259, 2003

5. Akimoto T, Muramatsu H, Takahashi M, et al: Rectal bleeding after hypo-fractionated radiotherapy for prostate cancer: Correlation between clinicaland dosimetric parameters and the incidence of grade 2 or worse rectalbleeding. Int J Radiat Oncol Biol Phys 60:1033-1039, 2004

6. Tsuji H, Yanagi T, Ishikawa H, et al: Hypofractionated radiotherapywith carbon ion beams for prostate cancer. Int J Radiat Oncol Biol Phys63:1153-1160, 2005

7. Higgins GS, McLaren DB, Kerr GR, et al: Outcome analysis of 300prostate cancer patients treated with neoadjuvant androgen depriva-tion and hypofractionated radiotherapy. Int J Radiat Oncol Biol Phys65:982-989, 2006

8. Soete G, Arcangeli S, De Meerleer G, et al: Phase II study of a four-weekhypofractionated external beam radiotherapy regimen for prostate can-cer: Report on acute toxicity. Radiother Oncol 80:78-81, 2006

9. Martin JM, Rosewall T, Bayley A, et al: Phase II trial of hypofractionatedimage-guided intensity-modulated radiotherapy for localized prostateadenocarcinoma. Int J Radiat Oncol Biol Phys 69:1084-1089, 2007

0. Ritter MA, Forman JD, Kupelian PA, et al: A phase I/II trial of dose-per-fraction escalation for prostate cancer. Int J Radiat Oncol Biol Phys69:S174, 2007

1. Lukka H, Hayter C, Julian JA, et al: Randomized trial comparing twofractionation schedules for patients with localized prostate cancer.J Clin Oncol 23:6132-6138, 2005

2. Yeoh EE, Holloway RH, Fraser RJ, et al: Hypofractionated versus con-ventionally fractionated radiation therapy for prostate carcinoma: Up-dated results of a phase III randomized trial. Int J Radiat Oncol BiolPhys 66:1072-1083, 2006

3. Pollack A, Hanlon AL, Horwitz EM, et al: Dosimetry and preliminaryacute toxicity in the first 100 men treated for prostate cancer on arandomized hypofractionation dose escalation trial. Int J Radiat OncolBiol Phys 64:518-526, 2006

4. Khoo VS, Dearnaley DP: Question of dose, fractionation and technique:Ingredients for testing hypofractionation in prostate cancer—TheCHHiP Trial. Clin Oncol (R Coll Radiol) 20:12-14, 2008

5. Bentzen SM, Ritter MA: The alpha/beta ratio for prostate cancer: Whatis it, really? Radiother Oncol 76:1-3, 2005

6. Madsen BL, Hsi RA, Pham HT, et al: Stereotactic hypofractionatedaccurate radiotherapy of the prostate (SHARP), 33.5 Gy in five fractionsfor localized disease: first clinical trial results. Int J Radiat Oncol BiolPhys 67:1099-1105, 2007

7. Pawlicki T, Cotrutz C, King C: Prostate cancer therapy with stereotacticbody radiation therapy. Front Radiat Ther Oncol 40:395-406, 2007

8. Tang C, Loblaw DA, Cheung P, et al: Condensing external beam radio-therapy to five fractions for low-risk localized prostate cancer. Earlyresults of pHART3. ASCO 2008 Genitourinary Cancer Symposium.

San Francisco, CA, February 14-16, 2008, p 148