WRK

Ut1wRRmdipelovr

F

A

1

hen and How Should Adjuvantadiation Be Used in Early Endometrial Cancer?

aled M. Alektiar, MD

The role of adjuvant radiation therapy (RT) in early endometrial cancer is no longer clearlydefined. The increased use of surgical lymph nodes staging and the perceived “lack ofsurvival advantage” with adjuvant pelvic RT leads to a wide array of conclusions on whoshould be treated and if so how. The purpose of this review is 2-fold: first, to determine thescientific validity of some of the conclusions drawn about the lack of survival impact ofadjuvant pelvic RT and, second, to determine which treatment option provides the besttherapeutic ratio. Overall survival may not be the ideal endpoint for early-stage endometrialcancer where death more often than not is because of causes other than endometrialcancer. Observation after hysterectomy may have the best morbidity profile, yet it may notbe the option with the best therapeutic ratio. Finding a suitable alternative such asintravaginal RT or using intensity-modulated radiation therapy may ultimately prove to bethe option with the best therapeutic ratio. The data learned from surgical lymph nodesstaging and from the 2 recent randomized trials on the role of pelvic RT in early endometrialcancer need not be ignored nor held as the final answer. Perhaps, rectal cancer should beused as an example of how an incremental use of individual adjuvant therapies, eachimpacting outcome little at a time, ended up improving overall survival. If we were to usethe current prevailing approach in endometrial cancer for rectal cancer, patients would stillbe treated with surgery alone.Semin Radiat Oncol 16:158-163 © 2006 Elsevier Inc. All rights reserved.

KEYWORDS radiation, intravaginal radiation, early stage endometrial cancer

jrstmiaRms

SRT(nssT

p until the early 1990s, the adjuvant treatment for earlystage endometrial cancer was straightforward. Most pa-

ients with less than half myometrial invasion and with gradeor 2 histology received intravaginal radiation therapy (RT),hereas those with deep invasion or grade 3 received pelvicT.1 The only area of controversy was whether intravaginalT should be added to patients receiving pelvic RT.2 Twoajor recent developments, however, have shaken the foun-ations of adjuvant RT in early endometrial cancer. First, the

ncreased use of surgical lymph node staging in many of theseatients has called into question the validity of using postop-rative pelvic RT for patients with pathologically negativeymph nodes.3-5 The other development was that the resultsf 2 prospective randomized trials reported no survival ad-antage with the addition of pelvic RT to surgery.6,7 Theeason these 2 developments rocked the foundations of ad-

rom the Department of Radiation Oncology, Memorial Sloan-KetteringCancer Center.

ddress reprint requests to Kaled M. Alektiar, MD, Department of RadiationOncology, Memorial Sloan-Kettering Cancer Center, New York, NY

e10021. E-mail: alektiak@mskcc.org

58 1053-4296/06/$-see front matter © 2006 Elsevier Inc. All rights reserved.doi:10.1016/j.semradonc.2006.02.004

uvant RT in endometrial cancer has to do with the unwar-anted conclusion that says that all forms of adjuvant RThould be abandoned in this disease. The aim of this review iso challenge this conclusion by first showing that there isore to survival benefit than we are led to believe, especially

n early endometrial cancer, and, second, to evaluate whethern alternative to conventional pelvic RT such as intravaginalT or pelvic intensity-modulated radiation therapy (IMRT)ay provide patients with a better therapeutic ratio than

urgery alone.

urvival Advantage of PelvicT in Early Endometrial Cancer

he Postoperative Radiation Therapy in Endometrial CancerPORTEC) trial randomized 715 patients after total abdomi-al hysterectomy and bilateral salpingo-oophorectomy to ob-ervation or pelvic RT.6 Patients included were those withtage I-B grades 2 and 3 and those with I-C grades 1 and 2.hose with I-B grade 1 and those with I-C grade 3 were

xcluded because it was thought that adjuvant RT was not

iityrRd.hpc

G(hwtdafTca4�Bnscdt

IOppct8Rc7hclndovpiRwGrec

ei

CArtoaf(cdoRtocher4wtpfasiSftscGfflaatRbb

RTvtotscvss

Early endometrial cancer 159

ndicated for the former, and most physicians would not omitt for the latter. No lymph node sampling was performed, andhe dose of pelvic RT was 46 Gy at 2 Gy per fraction. At 5ears, there was a statistically significant difference in theates of vaginal/pelvic recurrence in favor of adjuvant pelvicT (14% v 4%, P � .001). Overall survival, however, was notifferent between the 2 groups (81% RT v 85% surgery, P �

31), and the complications with pelvic RT were significantlyigher (25% v 6%, P � .0001). In addition, many of theatients who relapsed locally after surgery alone were suc-essfully salvaged with subsequent definitive RT.

The second randomized trial was Gynecology Oncologyroup (GOG) #99. There were 190 patients with stage IB-IIB

grades 1 to 3) disease who all underwent total abdominalysterectomy and bilateral salpingo-oophorectomy, pelvicashing, and pelvic/para-aortic lymph nodes sampling and

hen were randomized to observation versus pelvic RT to aose of 50.4 Gy at 1.8 Gy per fraction.7 At 2 years, there wasstatistically significant difference in the rates of relapse in

avor of the adjuvant pelvic RT arm (3% v 12%, P � .007).he 2-year estimated incidence of isolated vaginal/pelvic re-urrence was 1.6% in the RT group and 7.4% in the surgery-lone group. There was, however, no significant difference in-year overall survival (92% RT v 86% with surgery alone, P

.557), but there were more complications with pelvic RT.ased on the results of these 2 trials, some have adopted aew paradigm for early endometrial cancer that says “surgicaltaging for all, adjuvant RT for none, investigate the role ofhemotherapy for some.” In this review, the focus will be onetermining the validity of omitting adjuvant RT based onhe data from these 2 randomized trials.

s Overall Survival a Realistic Endpoint?verall survival is considered the gold standard primary end-oint in many randomized trials in oncology. Many of theatients with early endometrial cancer have competingauses of death other than endometrial cancer such as hyper-ension, diabetes, and obesity. In the PORTEC trial, the-year actuarial rates of intercurrent death were 19.7% in theT arm and 15.6% in the surgery-alone arm.8 Endometrialancer–related deaths in comparison were only 9.6% and.5%, respectively. Similarly, in GOG #99, approximatelyalf of the deaths were from causes other than endometrialancer or treatment (surgery alone, 19/36; RT, 15/30). Thised the authors of GOG #99 to write the following: “With thisumber of intercurrent deaths in both arms, even if RT re-uces the risk of endometrial cancer-related deaths, the sizef this trial is not adequate to reliably detect an overall sur-ival difference.”7 And that is why overall survival was not therimary endpoint in GOG # 99 but rather the disease-free

nterval, which was significantly different in favor of adjuvantT over surgery alone.7 In the PORTEC trial, the sample sizeas larger, but the number of events is perhaps smaller thanOG #99 because those patients with the highest risk of

ecurrence (stage I-C grade 3 and stage II patients) werexcluded from the study. Therefore, it is not unreasonable to

onclude that neither PORTEC nor GOG #99 are large e

nough to conclusively show whether adjuvant pelvic RTmpact overall survival.

omplications of Pelvic RTgreat deal of emphasis has been placed on the increased

isk of complications from adjuvant pelvic RT as reported inhe PORTEC trial, yet an in-depth analysis shows that somef the difference could be explained away. The 5-year actu-rial rates of complications (grade 1-4) in the trial were 26%or the RT arm compared with 4% in the surgery-alone armP � .0001). However, when grade 1 toxicity (68% of allomplications) was excluded, the corresponding ratesropped to 17% and 4%, respectively. Furthermore, the ratef grade 3 to 4 complications (significant) was only 3% in theT group.9 Almost 50% of all complications resolved over

ime.9 This change was mainly seen in patients with grade 1r 2 complications who represent most patients with compli-ations. For those few patients with grade 3 complications,owever, the symptoms never fully recovered as one mightxpect from high-grade complications. The 5-year actuarialate of complications was 21% for those treated with the-field box technique compared with 30% for those treatedith anterior-posterior/posterior-anterior field and 36% for

hose treated with 3-field approach. 9 This difference in com -lication rates approaches statistical significance (P � 006),

avoring the 4-field approach, which is also the preferredpproach for postoperative pelvic RT in the United States. Ithould not be surprising to most that there is great disparityn the reporting of toxicity between surgery and radiation.urgeons do not consider a permanent incision extendingrom above the umbilicus to the pubis a complication ofherapy, whereas a transient skin erythema from RT is con-idered grade 1 or 2 toxicity. Such disparity in reportingould not be highlighted any better than what the authors ofOG # 99 stated: “It must be pointed out that the mechanism

or reporting toxicity is slightly different for the two arms. Aorm used to report acute effects during radiation was col-ected for patients assigned to receive pelvic radiation ther-py, while standard follow-up forms were used to captureny other adverse events for all patients.”7 Therefore, it is safeo conclude that in the 2 randomized trials adjuvant pelvicT did increase the risk of complications over surgery alone,ut the magnitude of difference is not as large as we are led toelieve.

T For Salvagehe majority of recurrences in the PORTEC trial6 were in theagina (73%) and salvage treatment (mainly definitive RT) inhose patients was often successful with a 2-year survival ratef 79%. Based on those excellent results, some concludedhat RT should be reserved for salvage instead of the adjuvantetting. Such interpretation, however, should be viewed withaution. Of the 715 patients in the PORTEC trial, 39 hadaginal recurrence, and of those 39 patients, 32 were in theurgery-alone arm.9 In essence, we are being asked to adoptalvage RT as being the best treatment approach for early

ndometrial cancer based on the outcome of 32 patients, of

wwlpttloyraa4s

AoOcocrRtOii

WTautcertniasv

ts

povmnrnladrtaygcpwsw(a.

whiysldm�rtsvoa(oi

WIowtIh1Sc

T

S

W

J

A

160 K.M. Alektiar

hom only 26 attained complete remission and only 20 of 26ere with no evidence of disease after an unspecified fol-

ow-up period. In addition, there is no mention of how thoseatients were treated and what type and grade of complica-ions they developed. There is clear evidence from the data inhe literature10-12 that definitive RT can successfully treat iso-ated vaginal recurrences, but what is also clear is that theutcome with longer follow-up is not as high as 79% at 2ears as reported by PORTEC. Rather, the 5-year survivalates range from 50% to 60%. Such survival rates are oftenchieved by the combined use of external-beam radiationnd brachytherapy including interstitial implants to a dose of0 Gy each (Table 1). Thus, it is not surprising that the rate ofignificant complications12,13 could be as high as 8% to 10%.

ltering the Therapeutic Ratiof RT in Early Endometrial Cancerbviously, the data from the PORTEC and GOG # 99 trials

annot be ignored. But rather than concluding that all formsf adjuvant RT should be abandoned in early endometrialancer, it is perhaps better to determine if the therapeuticatio of adjuvant RT could be improved. If we omit adjuvantT, it would decrease the risk of complications yet increase

he risk of recurrence, thus worsening the therapeutic ratio.n the other hand, substituting conventional pelvic RT with

ntravaginal RT (IVRT) when indicated or pelvic IMRT maymprove the therapeutic ratio.

ho Can Have Surgery Alone?he PORTEC and GOG # 99 were set out to determine ifdjuvant pelvic RT is needed in early-stage endometrial, butnfortunately neither was large enough to answer the ques-ion of overall survival. Both trials, however, showed statisti-ally significant differences in terms of relapse. It is alwaysnticing to try to determine if there are subsets of patients inandomized trials who did or did not benefit from adjuvantherapy. Such an exercise is undertaken because of the sig-ificant clinical implications for everyday practice; thus, its

mportance cannot be overstated. Yet, findings from subsetnalysis should be accepted with caution. If a trial showed atatistically significant reduction in recurrence with an adju-

able 1 RT as Salvage for Vaginal Recurrence

AuthorsNo ofpts

5-YearLC

5-YearOS

PelvicRT BRT

ears etal10

45 54 44 50 Gy 40 Gy

ylie etal11

58 65 53 45 Gy 40 Gy

hingran etal12

57 75 42 50 Gy 40 Gy

bbreviations: Pts, patients; LC, local control; OS, overall survival;RT, radiation therapy; BRT,brachytherapy.

ant treatment, the burden of proof is on those who say that g

he adjuvant treatment should be omitted in certain patientubsets and not the other way around.

So the question becomes whether or not there is a group ofatients with very low risk of relapse in which RT could bemitted. The answer to that question relates to the risk ofaginal versus pelvic versus distant relapse. Distant relapse isore of a question of whether adjuvant systemic therapy iseeded, which is beyond the scope of our discussion. Pelvicelapse, in the current environment of routine surgical lymphode staging, relates to whether pathologically negative

ymph node sampling is enough evidence to omit pelvic RT,n issue that we will address later. Therefore, what it boilsown to when deciding on whether to give adjuvant RT is theisk of vaginal recurrence. The rate of vaginal relapse is nothe same for all early-stage endometrial cancer. In the surgerylone arm of the PORTEC trial, the risk of vaginal relapse at 5ears was 5% for I-B grade 2, 14% for I-B grade 3, 10% for I-Crade 1, and 13% for I-C grade 2 (Creutzberg CL, personalommunication, May 2004). Mariani and coworkers14 re-orted on 632 patients with stage I endometrial cancer, ofhom 508 were treated with surgery alone (152/508 were

tage I-A). The 5-year rate of vaginal relapse was 2% for thoseith grade 1 or 2 compared with 11% for those with grade 3

P � .0001). The presence of lymphovascular invasion (LVI)lso increased the vaginal relapse rate from 3% to 7% (P �02).

Another factor that is often ignored when deciding onhether adjuvant RT is needed is the patient’s age. Thereave been several reports in the literature about the negative

mpact of advancing age on outcome. In a multivariate anal-sis, Scholten and coworkers15 in a recent update of PORTEChowed that age �60 years was an independent predictor ofocoregional relapse (hazard ratio (HR), 3.4; P � .0005) andeath from endometrial cancer (HR, 2.6; P � .003). Further-ore, for patient’s �60 years with either grade 3 histology or50% myometrial invasion, the 10-year risk of locoregional

elapse (most were vaginal) increased from 4.6% for thosereated with adjuvant RT to 23.1% for those treated withurgery alone. Therefore, when deciding on whether adju-ant RT is needed, it is important to take into account the riskf vaginal relapse according to age, grade, depth of invasion,nd LVI. At Memorial Sloan-Kettering Cancer CenterMSKCC), we recommend observation to all stage I-A grade 1r 2 patients and offer it to most stage I-B grade 1 or 2 patientsf they are �60 years of age and have no LVI.

ho Can Be Treated With IVRT Alone?f IVRT alone were to improve the therapeutic ratio over thatf pelvic RT, its morbidity profile needs to be better butithout compromising either vaginal or pelvic control. Even

hough there has never been a randomized trial comparingVRT alone to pelvic RT alone, most data in the literature onigh-dose rate (HDR) IVRT shows a very favorable rate (0%-%) of significant complications16-19 as shown in Table 2.uch low rates of complications were obtained by payinglose attention to total dose, dose per fraction, length of va-

ina treated, and prescription depth. Sorbe and Smeades20

rt1p2mlo.wGWIe.dtftgmuG3

0ifnwtdacni

tghapslo

iqswpgtbn

PIpeTdwcclphnrsp

m

T

MPAJ

TA

MPAAS

T

MPAA

*

TS

CAHS

Early endometrial cancer 161

eported on 404 patients with stage I endometrial cancerreated with HDR IVRT. The rate of late complications was5.8% (6.9% were grade �2), which correlated with the doseer fraction (4.5-9 Gy). Onsrud and coworkers21 reported on17 patients also treated with postoperative HDR IVRT. Onultivariate analysis, dose to the vaginal mucosa, which re-

ates to applicator diameter and prescription depth, was thenly independent predictor of vaginal complications (P �007). In a recent trial by Sorbe and coworkers,22 290 patientsith stage IA-IB endometrial cancer were randomized to 15y at 2.5 Gy per fractions versus 30 Gy at 5 Gy per fraction.hen the length of vagina was measured at baseline before

VRT and at 5-years after IVRT, there was a significant differ-nce in the amount of vaginal shortening (2.1 cm, 25%; P �000001) for those treated to a total dose of 30 Gy but noifference for those treated to total of 15 Gy. At MSKCC, ourotal dose of HDR IVRT alone is 18 to 21 Gy given in 3ractions at 1- to 2-week intervals. The length of vaginareated ranges from 4 to 7 cm, depending on tumor stage andrade. The dose is prescribed to 0.5 cm from the vaginalucosa and to decrease the dose to vaginal surface; whensing small cylinders, the dose per fraction is lowered from 7y to 6 Gy if the diameter of the cylinder used is less thancm.Vaginal recurrence with HDR IVRT alone (range,

%-1.8%) is similar to that with pelvic RT16-18,23,24 as shownn Table 3. Pelvic recurrence when using IVRT alone is aunction of tumor stage/grade and whether surgical lymphode staging was performed. Stage IB grade 1 or 2 patientsith or without surgical lymph node sampling could be

reated with IVRT alone. The risk of pelvic recurrence fromifferent series in the literature16,17,23,25 ranges from 0% to 2%s shown in Table 4. Stage IB grade 3 and IC patients areonsidered at high risk for having positive pelvic lymphodes; therefore, whether or not surgical lymph nodes stag-

ng was performed is important. In the Aadlers randomized

able 2 Significant Complications With HDR IVRT Alone

Author Year No of pts%

Complication

cLeod et al16 1998 143 0etereit et al17 1999 191 0lektiar et al18 2005 382 1olly et al19 2005 50 0

able 3 Rate of Vaginal Recurrence With HDR Intravaginal RTlone

Authors Year StageNo ofpts

VaginalRecurrence

(%)

cLeod et al16 1998 IB-IIB 133 2 (1.5)etereit et al17 1999 IB-C 171 0 (0)nderson et al23 2000 IB-C 102 1 (1)lektiar et al18 2005 IB-IIB 382 7 (1.8)

olhjem et al24 2005 IB-IC 94 0 (0)

*

rial26 of IVRT alone versus pelvic RT � IVRT, those with IBrade 3 and stage IC without lymph nodes staging had aigher rate of locoregional relapse when treated with IVRTlone, indicating that in the absence of surgical staging thoseatients need pelvic RT. For stage IB grade 3 and IC withurgical lymph node staging, IVRT alone seems to provideow-enough pelvic recurrence (range, 0%-2.4%) to justifymitting pelvic RT23,24,27,28 as shown in Table 5.Despite the surge in the use of surgical lymph node staging

n early-stage endometrial cancer, what constitutes an ade-uate staging is not well defined. This is an important con-ideration because we are being asked to drop pelvic RT, aell-established treatment modality, in favor of a new ap-roach with a shorter track record. Most gynecologic oncolo-ists would agree that sampling of at least 10 lymph nodes ishe minimum number required. In addition, sampling ofoth right and left lymphatic stations in the obturator, inter-al/external/common iliacs, and para-aortic is required.

elvic IMRTntravaginal RT is a reasonable alternative to conventionalelvic RT in most patients with early endometrial cancer,specially for those who had surgical lymph nodes staging.here are some patients, however, who are not good candi-ates for surgical staging or with poor prognostic factors inhom pelvic RT is generally recommended. The former in-

lude obese patients or those with other morbidity that pre-ludes comprehensive surgical staging, thus having only fewymph nodes removed or none at all. The latter relate toresentations in which the risk of pelvic relapse is consideredigh even in the presence of pathologically negative lymphodes. An example of this is extensive LVI in which the risk ofelapse in the lymph nodes is still high or in patients withtage II-B with deep stromal invasion in whom the risk ofarametrial relapse is high as well.In such cases, replacing conventional pelvic RT with IMRTight prove useful. There have been several publications on

able 4 Vaginal IVRT Alone* for IB Grade 1 or 2

Author YearNo ofpts

PelvicRecurrence

(%)

acLeod et al16 1998 83 2etereit et al17 1999 153 1nderson et al23 2000 62 2lektiar et al25 2002 233 2

Some patients did have surgical lymph nodes sampling.

able 5 Pelvic Relapse After IVRT Alone and Lymph Nodesampling for IB Grade 3-IC

Authors Year No of pts n (%)

hadha et al27 1999 38 0 (0)nderson et al23* 2000 44 1 (2.2)orowitz et al28 2002 81 2 (2.4)olhjem et al24 2005 60 0 (0)

Depth defined as inner, middle, outer.

tpd4ac6vc1(cPrspvwndsoa

TBpsbisvtrfc2p

lwtFgTrtt�pmswgitpcpwwwcr9TawplnnsohcahmstItimfimwpa

R

TE

III

*

TE

III

*

162 K.M. Alektiar

he dosimetric advantages29,30 of pelvic IMRT, but more im-ortantly there are data to show that the morbidity could beecreased with IMRT. Mundt and coworkers31 reported on0 patients with gynecologic cancers treated with postoper-tive pelvic IMRT compared with 30 patients treated withonventional pelvic RT. Grade 2 acute GI toxicity was seen in0% with IMRT compared with 91% with conventional pel-ic RT (P � .002). In a subsequent publication, Mundt andoworkers32 reported that chronic GI toxicity was seen in1.1% with IMRT compared with 50% with conventional RTP � .001). On multivariate analysis, IMRT was still signifi-ant (relative ratio, 0.16; 95% confidence interval, 0.04-0.67;� .01). An added benefit of pelvic IMRT has been the

eduction in hematologic toxicity. Brixey and coworkers33

howed a reduction in hematologic toxicity especially foratients receiving pelvic RT and chemotherapy (CT). Con-entional pelvic RT � CT patients developed lower medianhite blood cells (2.8 v 3.6 �g/dL, P � .05) and absoluteeutrophil count (ANC) (1,874 v 2,669, P � .04) nadirs thanid IMRT � CT patients. Therefore, it could be argued thatome of the shortcomings of conventional pelvic RT could bevercome by IMRT, thus improving the therapeutic ratio ofdjuvant RT in early stage endometrial cancer.

reatment Recommendationsased on the available data in the literature, conventionalelvic RT is perhaps overkill for some patients with early-tage endometrial cancer, especially those with a large num-er of pathologically negative lymph nodes at surgical stag-

ng. It is also important to recognize that a negative surgicaltaging should not be translated into recommending obser-ation for all patients. A better approach would be to tailorhe adjuvant RT based on the risk of vaginal versus pelvicelapse. The current treatment recommendations at MSKCCor early-stage endometrial cancer (serous and clear-cell ex-luded) are shown in Tables 6 and 7. For stage IA grade 1 or, observation is recommended because the risk of vaginal orelvic relapse is very low. For stage IA grade 3, because of the

able 6 MSKCC Treatment Recommendations for Early-Stagendometrial Cancer Without Lymph Nodes Surgical Staging

Stage Grade 1 Grade 2 Grade 3

A None None IVRT or noneB IVRT or none IVRT or none Pelvic RT*C-IIB (occult) Pelvic RT* Pelvic RT* Pelvic RT*

Consider pelvic IMRT.

able 7 MSKCC Treatment Recommendations for Early-Stagendometrial Cancer With Lymph Nodes Surgical Staging

Stage Grade 1 Grade 2 Grade 3

A None None IVRT or noneB IVRT or none IVRT or none IVRTC-IIB* (occult) IVRT IVRT IVRT

IVRT alone if the stromal invasion was superficial otherwise pelvic

IMRT.

imited amount of data in the literature, it is hard to knowhat the best treatment recommendation is. At MSKCC, ei-

her IVRT or observation is recommended for those patients.or stage IB grade 1 or 2, the risk of pelvic relapse in thisroup of patients is considered too low to justify pelvic RT.herefore, the treatment recommendation is based on theisk of vaginal relapse, which seems to be related to the pa-ient’s age and LVI. At MSKCC, we recommend IVRT forhose �60 years old or with LVI. For those without LVI and

60 years old, we offer them either observation or IVRT. Foratients with stage IB grade 3 and IC, the treatment recom-endation is mainly based on whether surgical lymph node

taging was performed. Stage IB grade 3 and IC all gradesithout surgical lymph node staging: Patients with stage IBrade 3 and IC, all grades without surgical lymph node stag-ng, have a high risk of pelvic recurrence and should bereated with pelvic RT. Whether IVRT should be added toelvic RT has been a topic of extensive debates. Greven andoworkers34 reviewed the experience of 2 institutions to com-are the outcome of the 2 approaches. In that study, thereere 270 patients with stage I to II endometrial cancer; 173ere treated with postoperative pelvic radiation alone and 97ith combination of intravaginal and pelvic radiation. The

orresponding 5-year pelvic control and disease-free survivalates of pelvic RT alone versus with IVRT were 96% versus3% (P � .32) and 88% versus 83% (P � .41), respectively.his study as well as others called into question whether theddition of vaginal radiation is needed. Currently at MSKCC,e recommend adjuvant pelvic IMRT alone for this group ofatients. For stage IB grade 3 and IC, all grades with surgical

ymph node staging, if the patients had at least 10 lymphodes removed from right and left obturator, internal/exter-al/common iliacs, and para-aortic, then IVRT alone is a rea-onable alternative to pelvic RT. At MSKCC, we do not rec-mmend observation for this group of patients unless theyave comorbidity that significantly increases their risk ofomplications. There is growing literature on using IVRTlone for patients with stage II disease.35-37 It is important,owever, to take into account the extent of cervical involve-ent when determining the type of RT to recommend. For

tage 11A, for patients with cervical gland involvement only,he treatment recommendations follow that of IB grade 3 toC. For occult 11B, for those with stromal involvement, thereatment recommendations depend on the extent of stromalnvasion. We generally ask our gynecologic pathologist to

easure the amount of cervical invasion, and if it was super-cial, we recommend IVRT alone. For those with deep stro-al invasion, we recommend adjuvant pelvic IMRT evenith negative lymph nodes because of the high risk ofarametrial recurrence. Perhaps, this subset of patients couldlso benefit the most from pelvic RT plus IVRT.

eferences1. Piver M, Hempling R: A prospective trial of post-operative vaginal

radium/cesium for grade 1-2 less than 50% myometrial invasion andpelvic radiation therapy for grade 3 or deep myometrial invasion insurgical stage I endometrial adenocarcinoma. Cancer 66:1133-8, 1990

2. Randall ME, Wilder J, Greven K, et al: Role of intracavitary cuff boost

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

Early endometrial cancer 163

after adjuvant external irradiation in early endometrial carcinoma. Int JRadiat Oncol Biol Phys 19:49-54, 1990

3. Orr JW Jr, Holimon JL, Orr PF: Stage I corpus cancer: is teletherapynecessary? Am J Obstet Gynecol 176:777-788, 1997

4. Mohan DS, Samuels MA, Selim MA, et al: Long-term outcomes oftherapeutic pelvic lymphadenectomy for stage I endometrial adenocar-cinoma. Gynecol Oncol 70:165-171, 1998

5. Straughn JM Jr, Huh WK, Kelly FJ, et al: Conservative management ofstage I endometrial carcinoma after surgical staging. Gynecol Oncol84:194-200, 2002

6. Creutzberg CL, van Putten WL, Koper PC, et al: Surgery and postop-erative radiotherapy versus surgery alone for patients with stage-1 en-dometrial carcinoma: Multicentre randomised trial. PORTEC StudyGroup. Post Operative Radiation Therapy in Endometrial Carcinoma.Lancet 355:1404-1411, 2000

7. Keys HM, Roberts JA, Brunetto VL, et al: A phase III trial of surgery withor without adjunctive external pelvic radiation therapy in intermediaterisk endometrial adenocarcinoma: A Gynecologic Oncology Groupstudy. Gynecol Oncol 92:744-751, 2004

8. Creutzberg CL, van Putten WL, Koper PC, et al: Survival after relapse inpatients with endometrial cancer: results from a randomized trial. Gy-necol Oncol 89:201-209, 2003

9. Creutzberg CL, van Putten WL, Koper PC, et al: The morbidity oftreatment for patients with stage I endometrial cancer: Results from arandomized trial. Int J Radiat Oncol Biol Phys 51:1246-1255, 2001

0. Sears J, Greven K. Hoen HM, et al: Prognostic factors and treatmentoutcome for patients with locally recurrent endometrial cancer. Cancer74:1303-1308, 1994

1. Wylie J, Irwin C, Pintilie M, et al: Results of radical radiotherapy forrecurrent endometrial cancer. Gynecol Oncol 77:66-72, 2000

2. Jhingran A, Burke TW, Eifel PJ: Definitive radiotherapy for patientswith isolated vaginal recurrence of endometrial carcinoma after hyster-ectomy. Int J Radiat Oncol Biol Phys 56:1366-1372, 2003

3. Lilie L, Lin LL, Perry W, et al: Definitive radiotherapy in the manage-ment of isolated vaginal recurrences of endometrial cancer. Int J RadiatOncol Biol Phys 63:500-504, 2005

4. Mariani A, Dowdy SC, Keeney GL, et al: Predictors of vaginal relapse instage I endometrial cancer. Gynecol Oncol 97:820-827, 2005

5. Scholten AN, van Putten WLJ, Beerman H, et al: Postoperative radio-therapy for stage 1 endometrial carcinoma: Long-term outcome of therandomized PORTEC trial with central pathology review. Int J RadiatOncol Biol Phys 63:834-838, 2005

6. MacLeod C, Fowler A, Duval P, et al: High-dose-rate brachytherapyalone post-hysterectomy for endometrial cancer. Int J Radiat Oncol BiolPhys 42:1033-1039, 1998

7. Petereit DG, Tannehill SP, Grosen EA, et al: Outpatient vaginal cuffbrachytherapy for endometrial cancer. Int J Gynecol Cancer 9:456-462, 1999

8. Alektiar KM, Venkatraman E, Chi DS, et al: Intravaginal brachytherapyalone for intermediate-risk endometrial cancer. Int J Radiat Oncol BiolPhys 62:111-117, 2005

9. Jolly S, Vargas C, Kumar T, et al: Vaginal brachytherapy alone: Analternative to adjuvant whole pelvis radiation for early stage endome-trial cancer. Gynecol Oncol 97:887-892, 2005

0. Sorbe BG, Smeds AC: Postoperative vaginal irradiation with high dose-rate afterloading technique in endometrial carcinoma stage I. Int J Ra-

diat Oncol Biol Phys 18:305-314, 1990

1. Onsrud M, Strickert T, Marthinsen AB: Late reactions after postopera-tive high-dose-rate intravaginal brachytherapy for endometrial cancer:A comparison of standardized and individualized target volumes. Int JRadiat Oncol Biol Phys 49:749-755, 2001

2. Sorbe B, Straumits A, Karlsson L: Intravaginal high-dose-rate brachy-therapy for stage I endometrial cancer: A randomized study of twodose-per-fraction levels. Int J Radiat Oncol Biol Phys 62:1385-1389,2005

3. Anderson JM, Stea B, Hallum AV, et al: High-dose-rate postoperativevaginal cuff irradiation alone for stage IB and IC endometrial cancer. IntJ Radiat Oncol Biol Phys 46:417-425, 2000

4. Solhjem MC, Petersen IA, Haddock MG: Vaginal brachytherapy aloneis sufficient adjuvant treatment of surgical stage I endometrial cancer.Int J Radiat Oncol Biol Phys 62:1379-1384, 2005

5. Alektiar KM, McKee A, Venkatraman E, et al: Intravaginal high-dose-rate brachytherapy for stage IB (FIGO Grade 1, 2) endometrial cancer.Int J Radiat Oncol Biol Phys 53:707-713, 2002

6. Aalders J, Abeler V, Kolstad P, et al: Postoperative external irradiationand prognostic parameters in stage I endometrial carcinoma: Clinicaland histopathologic study of 540 patients. Obstet Gynecol 56:419-427,1980

7. Chadha M, Nanavati PJ, Liu P, et al: Patterns of failure in endometrialcarcinoma stage IB grade 3 and IC patients treated with postoperativevaginal vault brachytherapy. Gynecol Oncol 75:103-107, 1999

8. Horowitz NS, Peters WA 3rd, Smith MR, et al: Adjuvant high dose ratevaginal brachytherapy as treatment of stage I and II endometrial carci-noma. Obstet Gynecol 99:235-240, 2002

9. Portelance L, Chao CL, Grigsby PW, et al: Intensity-modulated radia-tion therapy (IMRT) reduces small bowel, rectum, and bladder doses inpatients with cervical cancer receiving pelvic and para-aortic irradia-tion. Int J Radiat Oncol Biol Phys 51:261-266, 2001

0. Roeske JC, Lujan A, Rotmensch J, et al: Intensity-modulated wholepelvic radiation therapy in patients with gynecologic malignancies. IntJ Radiat Oncol Biol Phys 48:1613-1621, 2000

1. Mundt AJ, Lujan AE, Rotmensch J, et al: Intensity-modulated wholepelvic radiotherapy in women with gynecologic malignancies. Int JRadiat Oncol Biol Phys 52:1330-1337, 2002

2. Mundt AJ, Mell LK, Roeske JC: Preliminary analysis of chronic gastro-intestinal toxicity in gynecology patients treated with intensity-modu-lated whole pelvic radiation therapy. Int J Radiat Oncol Biol Phys 56:1354-1360, 2003

3. Brixey CJ, Roeske JC, Lujan AE, et al: Impact of intensity-modulatedradiotherapy on acute hematologic toxicity in women with gynecologicmalignancies. Int J Radiat Oncol Biol Phys 54:1388-1396, 2002

4. Greven KM, D’Agostino RB Jr, Lanciano RM, et al: Is there a role for abrachytherapy vaginal cuff boost in the adjuvant management of pa-tients with uterine-confined endometrial cancer? Int J Radiat Oncol BiolPhys 42:101-104, 1998

5. Fanning J: Long-term survival of intermediate risk endometrial cancer(stage IG3, IC, II) treated with full lymphadenectomy and brachyther-apy without teletherapy. Gynecol Oncol 82:371-374, 2001

6. Ng TY, Nicklin JL, Perrin LC, et al: Postoperative vaginal vault brachy-therapy for node-negative stage II (occult) endometrial carcinoma. Gy-necol Oncol 81:193-195, 2001

7. Rittenberg PV, Lotocki RJ, Heywood MS, et al: Stage II endometrialcarcinoma: Limiting post-operative radiotherapy to the vaginal vault in

node-negative tumors. Gynecol Oncol 98:434-438, 2005