- Email: [email protected]
Controversies in the management of small cell lung cancer: thoracic radiotherapy in limited disease
Lung Cancer, 9 (1993) 253-264 0 1993 Elsevier !kientific Publishers Ireland Ltd. All rights
253
reserved0169-5002BMO6.00
LUNG 00143
Controversies in the management of small cell lung cancer: thoracic radiotherapy in limited disease Rodrigo Department
Arriagada”,
of'Radiation Oncology,
Jean-Pierre
Pignonb and Thierry
Le Chevalier’
bMedical Statistics and “Medical Oncology, Institut Gustave- Roussy (IGR), Mllejuif, France
Summary The review of randomized trials shows that thoracic radiotherapy combined with chemotherapy decreases the risk of thoracic recurrences and provides a gain in long-term survival of approximately 6% in limited disease when compared to chemotherapy alone. Thoracic radiotherapy and combined radiochemotherapy approaches are composite parameters and the analysis of their effects can be equivocal. The purpose of this paper is to discuss such parameters, namely treatment toxicity, volumes to be treated, total tumor dose, fraction size, timing of radiochemotherapy approaches and the quality of radiotherapy. The analysis of these parameters permits a delineation of questions to be tested in future clinical trials.
Key words: Small cell lung cancer; Limited disease; Thoracic radiotherapy and chemotherapy
radiotherapy;
Combined
With each step forward, with each problem which we solve, we not only discover new and unsolved problems, but we also discover that where we believed that we were standing on firm and safe ground, all things are, in truth, insecure and in a state of flux
KARL R. POPPER Introduction
Thoracic radiotherapy (RT) was the main treatment of small cell lung cancer (SCLC) before the 1970s; long-term survival was poor (below 5% at 5 years). From then on, Correspondence to: Dr. R. Arriagada, Desmoulins, 94805 Villejuif, France.
Dept. of Radiation
Oncology, Institut Gustave-Roussy,
rue Camille
,rCdwaqloypt?J awn[oA paDnpaJ,aq~ u! 88~ 01 paJoduro3UJJB,AdeJaqlo!pwauxn[oAap!M, aql u! %ZE .wM awJ acwaJJnDaJ ~peJoq~~J~u![[t?JaAo aqL *[oz] IDafqnss!qluo papnpuo3 SI?M‘swayEd [6[ 8u!pn[3u! ‘[y-11 paz!uropueJ auo @IO *sa!pnis aA!yadsoJ~aJ 6q palJoddns aJo3aq lc[uo s!uogepuauu.uo~aJsfq~‘JaAaMoH $JEW aaJ3~13z-s*[e ~I!M ‘ICdoJaqlowaqD pauyap w aurn[oAJownl [eu$i!Jo aql $t?aJl 01 s![E[‘or]uogt2puauuuo9aJpJt?puels ayJ_ auqod loum$ pampa snsm
aurnIo.4 ~oumj ~a@~0
.Mo[aqpassnwp aq [[!MLayI put!suoyeJa)u! asay II!aua]Joduq [I?!~~JDJO aJt!sJalawk?Jed luatuloaJ1 Jaylo ‘JaAaMoH asanssyasoy uo IDa33a D!xolaA!pa[aue ~I!M pali?posst? aJe qa!q~ up!qnJoxop put!uph_uoa[qSE q3ns s%Jp3ouo!geJ~syupe aq)p!oncogJa3osaqp[noM $!'snql fawn[oA ayl u!qy~ pnwo[ sanssg aql u! Q.qEur pahlasqo uaaq aAeq asaqL *suogt?a pa,wpeJJ!. -![dwo~ihqua)oaJq)-a3![ uaAa pun aJaAasaDnpoJd uw ~a put!~330 suogewqwo3 walxa put2Ll!suqu!w30 t2ap! Jea[DIsaaaq 01 I[n9gyps!I!puti2 snoauai?oJajaq 1cJaAaJe dl!~!xol asoyhi 6%alEJlsynadt? uo SlJOdaJ‘~[a)wrn)Jo3u~ .asaas!p ~ut?u%y?wa$t?cNpt?Ja 01 s!UI!I? -Jay]hI0 U! JO13E3%U!@IU[aqj Uaaq ShM[" aAE?qSanSS!] [OUUOU UO Spa&a 3!XOJ 'h$!D!XOl [t?q)a[u! ast?aJDu! UB s!aJaql3! u!dynadeJaq~ ous!aJaq~leq$pu!wu! auJoq aq p[noqs11
Xdtrmyo!per aplr.~oqjJOasn aq) II! W~~AO.~UO~
yt2!~1 aJnln3u! palsa)aq 01 suoysanb aleaugap0) ldwa)leUB u! SJalaUJt?Jt?d asaq)[[I? ssn%!p 0) s!JadEd s!ql30asodJndaqL .dJaA![ap 6doJaqlo!peJ30 Qynbaql(i?)puvsaq~ooJdde[tyuanbas-uou SnsJaA[guanbas 0) f/idEJaqlo!pw ale[SnsJaAA[Jt?a (a)fansUO9DEJ3(p)taSOpJOUIn~["]Oj (3)fpa$EaJl aq 01 w.un[oA (q)tL~p!xol luauI)saJ) (B) :~uno~~eOIU!uayE1aq 01 aAt?qspadst? [t?$uaun?pun3 uaAag .[woA!nbaaqUKI s$Da33aJ!aql3o s!d[t?u~aq~ pun SJalaUIEJI?dal!sodUIo3 aJt?ddl?Jaq~OUIaqDO!pEJ paU!qUIODpUE La D!%?JOqL 'Sa[npaq%UO!~l?U!qrUOD JO3 ih!UU~ [wxydoue auyapola[qEun aJaM SfSd[i?Ut?-E)atU aqlu!slvpaqlq~fMapt?wsuosp~dwo~paJ -!pu!aqL .aseas!p pa~!uu[30lUaI.Uaii!EUt?w aql u! saqDeoJddeAdEJaq$OUIaqDO!pt?J pau!qwoD 30 asn [I?JaUa% aq$ slJoddns['I?A!AJns [[T?JaAo 30 sI.uJa~ uf lyauaq pa~t?J~suouIap OS aqL *[v] aDuaJa3uo3s!qlu! pa$uasaJds[t?p$ pazyopuw papuno3uoDun E[ 30 s!s@ut?-"law aA!suaqaJduIoD B dq paurryuo:, dpua3aJ SBM [vA!hlnsurral-8uo[ u! u@ sty1 *[EC‘6 ‘[I asEas!p pa~!uu[u! 8()-s I([alt?w!xoJdde3o[eA!Nns uua@o[u! u$!t! pas!uJoJdpu?i?sa~uaJ -Jn3aJD!DeJOql30?fS!J aq$ paWaJ3ap La +?JOq$ It?ql paMOqS S[l?!J) paZ!UIOpU&?J30 Ma!AaJ aqL .uo!ss!waJ ala[dwoDJaUe Jn33o 01 panuyuo3 sawaJJn3aJIsay ~0s uttq)aJ0t.u SE wa[qoJd ~0fa1.U t2pau!swaJ [o~)uo9[wo[ y3eJoql ‘auo[e~3 qI!M pa]eaJlsluagt?d30sapas lSaqaql uy uaAa 3vq3luap!AaaweDaq 1!‘JaAaMOH *(sJl?ad g $I?%8 6[alt?uqxoJddt?)[eA!"ns ruJa@uo[ u! uyG3 aleraporut!aAa!qDE01 a[qt? ~I?M~3 ‘auo[t? pasn uaqM .&x01 palt?[aJ -1uawwaJ) pawaJw_q 6[~ut?c~@&spua ‘kre3! ‘lyauaqt?paJaJ0 6[aJt2q ~a [EDO[leq)Bu! -uy[~ sa[npaqDs cynadEJaq$J!aq$u! asn s$!papJeDs!pswea$ ,GIE&~%uau@aJ uo!y?u!qwo:, U! _La 3!3l?JOq$ 30 UO!Sn[DU! aq$ JaAO kSJaAOJ$UODI?0) aS!JaA& LD qyM pap[a$ Salt?J aSUOdSaJq&q aqL g13S303uaur)EaJl aql-ro3 pasnIC[ap!Maqo~u&aq(~3)ddeJaq~otuaq3
255 arm; this difference was not significant. On the other hand, three retrospective studies including a total of 159 patients [25,36,43] showed that the use of inadequate portals could cause a two- to three-fold increase in intrathoracic recurrences. This significant adverse effect could be related to major protocol variations as the recurrence rates were over 50%. Our own data [8] showed no increase in the local relapse rate in the event of inadequate coverage - strictly defined - of the initial tumor: 36% versus 33% in the case of adequate coverage. This finding agrees with the results of the above-mentioned randomized trial, and it would suggest that tumor shrinkage after chemotherapy is sufficient to allow the tumor to be completely encompassed by ‘inadequate’ fields. We should not hastily conclude, however, that small fields should be recommended for the treatment of these patients. Retrospective series clearly indicate that very limited fields do not improve local control compared to treatments with chemotherapy alone. Three further arguments should be taken into account: (1) the difficulties encountered when attempting to define what adequate coverage means, (2) the relatively small number of patients included in these studies, and (3) the fact that the most frequent site of relapse was found inside the fields when the irradiation encompassed the original tumor volume [8]. This finding is possibly indicative of an insufficient dose level [15, 341. The amount of the surrounding lung which should be treated after chemotherapy is important inasmuch as it is closely linked to treatment toxicity and treatment failure. Further randomized and a priori well-defined studies, using strict definitions of original and reduced tumor volumes specified by the individual investigators, are necessary to settle this issue. Low versus moderate versus high total radiation dose The optimal radiation dose level is still unknown. It is worthwhile to point out that only one randomized trial has been conducted on this subject [15]. The dose levels compared were low versus moderate: 25 Gy in 10 fractions over 2 weeks and 37.5 Gy in 15 fractions over 3 weeks. The authors concluded that the use of a higher dose seemed to delay rather than prevent thoracic recurrence. All other data are only provided by retrospective studies; a summary is shown in Table 1. Clearly, doses below 40 Gy are unable to control local disease, but nothing can be stated about the effect of moderate (40-50 Gy) versus high dose (> = 60 Gy) radiotherapy. It is also interesting to observe the wide variation in the local
TABLE 1 Total tumor dose delivered by thoracic radiotherapy in SCLC and overall local failure” Tumor dose (Gy)
Local failure (96)
<40 40 4P 50 360
79-100 2343 16-39 25-42 4-25
“Modified from Turrisi [41] summarizing retrospective data from eight centers. bIncludes accelerated fractionation.
256 failure rate at each dose level. This is probably related to different definitions of local failure in each center: (i) recurrence after complete remission versus overall local failure; (ii) overall failure versus first site of failure; (iii) methodology used to report the results: total, censored or competing event approaches [5]. Our own data [5,8] did not show a significant difference in long-term local control in four consecutive alternating radiochemotherapy protocols which delivered a total radiation dose from 45 Gy up to 65 Gy. However, we should underline that this was also a retrospective analysis based on a limited number of patients and that radiotherapy was delivered in 3 split courses. There is certainly room for a randomized trial testing a moderate versus a high total radiation dose, for example 40 Gy versus 65 Gy or their biological equivalent in non-conventional fractionation. Conventional fractionation versus unconventional fractionation Conventional fractionation, defined as 5 fractions of 2 Gy per week, allows a total radiation dose as high as 70 Gy to be delivered in a limited mediastinal volume. The fraction size used in the 13 randomized trials included in the meta-analysis [4] varied between 2 and 4 Gy; a large fraction size could be a possible explanation for increased toxicity reported in some of these trials. The fraction size can be reduced and repeated a few hours later on the same day. This is called hyperfractionation, in which case, the treatment time remains the same but the total dose is slightly higher. The RTOG has conducted trials in non-SCLC using fractions of 1.2 Gy twice daily and increasing the total dose up to 79 Gy in a limited volume [14]. The rationale for the use of this type of schedule has been extensively developed in the relevant literature [40,44]. It is based on the premise that healthy normal tissue can be relatively spared while higher doses of RT can be delivered to tumor cells. This approach can be particularly interesting in ‘pure’ SCLC, as demonstrated by Mitchell et al. [27] with the survival curves obtained from cell cultures. Looney et al. [23] studied in the rat hepatoma 3924A the possibility of integrating multifractioned radiotherapy in an alternating RT-CT schedule and concluded in terms of a probable benefit for cure. Some clinical studies have been conducted in limited SCLC [12, 18, 19, 421. The results are still preliminary, but are encouraging. In Table 2 the series of Turrisi and Johnson are summarized, in which accelerated fractionation (1.5 Gy twice daily) was used allowing 45 Gy to be delivered in 3 weeks with the concurrent administration of etoposide and cisplatin. TABLE 2 Pilot studies on concurrent RT and CT in limited SCLC Author
n
CT
Dose RT MS (Gy/fraction) (weeks)
2-yr S (%)
McCracken (26) Murray (29) Kwiatkowski (21) Turrisi (42) Johnson (19)
156 IO 50 32 38
VCR-VP-CDDP CAVNP-CDDP CYC-VP-CDDP VP-CDDP VP-CDDP
45125 30/10 50125 45130” 45/3p
45 32 ? 51 65
MS: median survival. “Bifractionated radiotherapy.
13 78 63 _ 113
257 Our own series (protocol 010) is described in column 5 of Table 4. Patients were treated by an alternating RT-CT schedule in which the first course of RT was given with accelerated hyperfractionation using reduced fields without treating the spinal cord. This first course consisted of a dose of 21 Gy given in fractions of 1.4 Gy three times daily. The results were not different from those observed with our previous protocols, but we registered a higher, albeit nonsignificant, incidence of death unrelated to cancer and probably related to treatment toxicity [5]. Two patients died with sepsis or infectious disease, one with heart failure, one with renal failure, and one with lung fibrosis [A. There are two possible explanations for this toxicity: (i) a higher initial dose of cisplatin (150 mg/m’ given over 5 days); (ii) a 4-h time gap between the 3 daily fractions. Currently, a time gap equal to or above 6 h is recommended to permit the repair of sublethal lesions in normal tissues [44]. It should be noted that small changes may have an unexpected impact on toxicity, and this makes it more difficult to evaluate different combined schedules. For instance, for the first 4 patients in this series we used a dose of 25 Gy in the first multifractionated course. Three developed early and unusually severe esophagitis. By simply reducing the dose to 21 Gy (16% of the multifractionated course dose and 6% of the total radiation dose) the rate and intensity of the mucosa reaction reverted to previous levels. Accelerated multifractionated radiotherapy is an interesting approach in the management of SCLC. However, its integration in combined schedules should be carefully planned and evaluated in prospective studies exploring toxicity, as small changes in treatment parameters may have an unexpected impact on early and late toxicity. Early versus late toxicity There are at least two theoretical reasons why early thoracic RT should continue to be included. The first is the appearance of chemoresistant cells and the second is tumor repopulation during treatment. If CT and RT have to be given sequentially and in full doses, the consequence of this protracted administration may be a high level of tumor repopulation [39,44]. Occult distant metastases will continue to proliferate if the adminisTABLE 3 Alternating RT-CT protocols Protocol
002 (1980-81)
004 (1982-84)
006 (1985-86)
010 (1987-88)
Number of patients Cyclophosphamide (mp/m2) Doxorubicin d I (mg/m*) VP16213 (mg/m*) Methotrexate (mg/m*) Cisplatin (mg/m’) Thoracic RT (Gy) PC1 (GY)~ Maintenance CT (number of cycles)
28 300x4 40 75x3 400 _
81 300x4 40 75x3
64 300x3 40 100x3
29 300x4 40 75x5
100 20-20-I 5 30 8
120 20-20-25 24” 6
30x5 2l”-20-20 24 0
15-15-15 30 I2
“Multifractionated RT given by reduced radiation fields. bFractions of 3 Gy. “Random,ized for complete responders.
258 tration of CT is delayed. For example, if the tumor doubling time is 30 days, a delay of 2 months would permit a four-fold increase in the volume of metastases [39]. RT cannot be delayed for too long because CT often only has a limited effect on bulky tumors and can even elicit the development of radioresistance. In view of the uncertainty regarding the extent of repopulation, the shortest delay between CT and RT is advisable. More recently, Murray et al. [28] have developed a mathematical model along these lines. The indirect comparisons effectuated in the SCLC meta-analysis [4] did not provide a definite answer to whether a preference should be given to early or late radiotherapy. However, three randomized studies have directly tested this question and their results are summarized in Table 3. The NCIC trial showed a significant difference 0, = 0.016) in favor of early radiotherapy [16]. In the CALGB trial [37] there was a slight difference in favor of late radiotherapy which may be explained by the non-delivery of full chemotherapy doses in the early radiotherapy arm due to increased acute toxicity. This finding suggests that initial full doses of chemotherapy may have an influence on survival [3]. The Danish trial [31] did not show a significant difference. For the time being, we are of the opinion that the use of early radiotherapy should be favored, but further research is needed to justify this attitude. Sequential versus nonsequential approaches
We have previously analyzed theoretical and clinical reasons to prefer concurrent or alternating to sequential radiotherapy approaches [ 1,2,39]. The main advantage of concurrent combinations is that it is possible to deliver RT early and without intervals between the modalities. The NC1 trial [ 1l] reported a very high incidence of complications and lethal toxicity in spite of the survival benefit obtained with the combined treatment. For a while, these results were a reference to avoid concurrent administration. However, the high rate of complications may have been due to the use of a moderate concentrated radiation schedule (fractions of 2.66 Gy 5 times a week) and the administration of doxorubicin. These two factors were avoided in the two concurrent arms of the CALGB and the NCIC trials and the toxicity seemed acceptable. In addition, some reports of pilot studies with similar schedules have demonstrated the feasibility of this kind of combination [21, 26, 291 with interesting mid-term results as summarized in Table 2. The alternating approach has been defined as: ‘combined modality therapy in which radiotherapy is given on days of the chemotherapy cycle in which no chemotherapy is TABLE 4 Early versus late radiotherapy in limited SCLC. Results of two randomized trials Group
n
Initial day of RT
2-yr Survival (%I
CALGB [37]
125 145 155 153 97 98
d, dti d, d 84 d, d 126
24 30 40 33 20 18
NCIC [I61 Aarhus [31]
259
-
SURVIVAL
-
RFS
Fig. 1. Overall survival and recurrence-free survival in 202 patients with limited stage SCLC treated by the alternating RT-CT schedules at the lnstitut Gustave-Roussy and cooperating centers (19Wl988).
administered, without any delay in chemotherapy beyond that which would occur if chemotherapy were being given alone’ [l]. It was developed to reconcile the needs for early administration of both agents and for sequential delivery of each modality without a long interruption [39]. This alternating regimen is consistent with experimental data reported by Looney et al. [24] using their rat hepatoma model; many experiments combining radiotherapy and cyclophosphamide permitted the definition of an optimal schedule: alternating courses of CT with 3 courses of RT with an interval of 7 days between each treatment modality. This treatment obtained a cure rate of 60% compared to 0% when the animals were treated by a single modality. The clinical investigation was developed simultaneously but independently in our Institute. The use of 3 split courses of RT in these schedules is open to criticism because of a decreased tumor effect due to tumor repopulation between the rest periods. This is true for tumors responding poorly to CT, but in SCLC interdigitated CT probably precludes tumor repopulation between RT courses. The details of the therapeutic schedule have been published previously [6, 221. Thoracic RT was started one week after completion of the second CT course and was given for 12 days with one rest week between both modalities. Three courses of RT alternated with CT which was continued up to 6 courses in the induction phase were followed by a complete assessment including fiber-optic bronchoscopy to verify complete remission. The first pilot trial was started in May 1980 at the IGR and later these studies were extended to four cooperating centers. The summarized description of these trials, which included 202 patients, is shown in Table 4. The overall and disease-free survival curves are shown in Fig. 1. In summary, alternating RT-CT schedules are an interesting area of research worth developing in the coming years. Their reproducibility has been proven in different French centers and they have been included in prospective trials. It must be recalled that this
260 approach was also used in one of the randomized trials which demonstrated a significant advantage with the combined approach [35]. Even if concurrent or alternating approaches promise a therapeutic gain, no randomized trial has been published comparing these combinations to sequential ones. One ongoing EORTC trial is comparing an alternating to a sequential combined radiochemotherapy schedule; the direct comparison between concurrent and alternating radiochemotherapy combinations is being explored by a French GETCB study. A very large number of patients will probable be needed to detect a significant difference between these schedules. Quality of radiotherapy delivery All previous considerations are only valid if the quality of the radiotherapy delivered is good. Poor quality radiotherapy may overshadow benefits and increase toxicity. Modem megavoltage radiotherapy standards include: (a) an unequivocal definition of the volumes to be treated; (b) an unequivocal definition of the tumor dose; (c) prechemotherapy simulator film and CT-scan based planning; (d) optimal beam arrangement to cover the previous defined volume and to protect critical organs; (e) computer dosimetry to describe dose distribution; (l) check films in the treatment machine to ensure treatment reproducibility. Of course, these standards should be used in the randomized trials aiming to define optimal treatment.
Discussion Progress in improving survival in limited SCLC has been slow during the last two decades. The change in standard treatment from radiotherapy alone to chemotherapy alone in the 1970s increased the 5-year survival rate from roughly 5% to 10%. The combination of radiotherapy and chemotherapy in the 1980s gave an additional benefit of 5% (Fig. 1). These moderate effects on survival should be tested in randomized trials with a large number of patients. Small or medium-sized trials can result in equivocal answers or yield a mixture of ‘positive’ and ‘negative’ results; most of these contradictions can be explained by statistical variations. Comprehensive meta-analyses using individual data may provide a solution to this problem [ 171. A case in point is the study which evaluated the role of thoracic radiotherapy in limited SCLC [4]. One of the main merits of this study was to contradict the contention that meta-analysis is not a feasible means of evaluating the treatment effect in lung cancer [30]. It also showed the advantages of meta-analyses based on individual data rather than on a review of the literature. This above-mentioned study concluded that thoracic RT gives a survival benefit of 5 to 6% at 3 years; this is the mean effect of 13 randomized trials comprising more than 2,000 patients. The next step will be to determine the optimal combined radiochemotherapy schedule. One question was answered, but many others are now being posed as the combined approaches are multiparametric. It is inherent to scientific knowledge that each step forward raises new problems that are of an ever-increasing depth [38]. In this paper we have discussed some aspects of what optimal radiotherapy and timing of optimal radiochemotherapy could consist of. We have not discussed what would be the most appropriate chemotherapy: type of drugs, drug doses, administration schedule, etc.
261 It is interesting to note that we are looking for moderate effects. However, hundreds of patients are needed to test differences between two treatment attitudes. It would be useless to compare very similar treatments considering the number of patients required; years of research would be lost with this type of strategy. Treatments with rather extreme differences should be investigated (e.g. moderate versus high radiation doses, conventional versus unconventional fractionations, sequential versus non sequential schedules) in order to test the validity of our hypotheses. This type of trial should be initiated fairly rapidly. The first trial testing the role of thoracic RT in SCLC was started in 1976 [32]. It took 15 years to collect sufficient data worldwide to provide an unequivocal answer. It is therefore in the interest of our patients that we start - or continue - to investigate the new questions raised, so as to optimize the management of SCLC with combined modalities by conducting large randomized trials on a massive scale during the next few years.
Acknowledgements The authors wish to thank the Centre Hospitalier Intercommunal de Creteil, the Hopital A. B&cl&e and the Fondation Bergonie participants in the IGR SCLC protocols, Mrs. M. Tarayre for data collecting, Mrs. G. Feris for secretarial assistance and Ms. L. Saint Ange for her assistance in revising the manuscript. The IGR SCLC 010 protocol was partially supported by the research grants INSERMKNAMTS no. 883063 and ECC no. ST-A000309.
References 1 Arriagada R, Bertino JR, Bleehen NM, et al. Consensus report on combined radiotherapy and chemotherapy
5 6 7 8 9 10 II
modalities in lung cancer. In: Arriagada R, I_e Chevalier, T, eds. Treatment Modalities in lung cancer. Karger, Basel: Antibiot Chemother 1988;41:232-241. Arriagada R, Cosset JM, Le Chevalier T, Tubiana M. The value of adjunctive radiotherapy when chemotherapy is the major curative method. 1nt.J Radiat Oncol Biol Phys 1990;19:1279-1284. Arriagada R, De The H, Le Chevalier T, et al. Limited small cell lung cancer: a possible prognostic impact of initial chemotherapy doses. Bull Cancer 1989;76:605+615. Arriagada R, lhde DC, Johnson DH, et al. Meta-analysis of randomized trials evaluating the role of thoracic radiotherapy in limited small cell lung carcinoma (abstract). Lung Cancer 1991; 7 (suppl):98. Arriagada R, Kramar A, Le Chevalier T, De Cremoux H, for the French Cancer Centers’ Lung Group. Competing events determining relapse-free survival in limited small-cell lung carcinoma. J Clin Oncol 1992; Arriagada R, Le Chevalier T, Baldereyou P, et al. Alternating radiotherapy and chemotherapy schedules in small cell lung cancer, limited disease. Int J Radiat Oncol Biol Phys 1985;11:1461-1467. Arriagada R, Pellae-Cosset B, Baldeyrou P, et al. Initial high dose chemotherapy and multifractionated radiotherapy in limited small cell lung cancer (abstract). Lung Cancer 1991; 7 (suppl): 159. Arriagada R, Pellae-Cosset B, Cueto Ladron de Guevara J, et al. Alternating radiotherapy and chemotherapy schedules in limited small cell lung cancer: analysis of local chest recurrences. Radiother Oncol 1991;20:91-98. Arriagada R, Pignon JP, Le Chevalier T. Thoracic radiotherapy in small cell lung cancer: rationale for timing and fractionation. Lung Cancer 1989;5:237-247. Bleehen NM. Radiotherapy for small cell lung cancer. Chest 1986;89:268S-2768. Bunn PA Jr., Lichter AS, Makuch RW, et al. Chemotherapy alone or chemotherapy with chest radiation therapy in limited stage small cell lung cancer: a prospective randomized trial. Ann Intern Med 1987;106:655662.
262 12 Choi NC, Propert K, Carey R. et al. Accelerated radiotherapy followed by chemotherapy for locally recurrent small-cell carcinoma of the lung. A phase II study of Cancer and Leukemia Group B. Int J Radiat Oncol Biol Phys 13:263-266. 13 Choi NC. Reassessment of the role of radiation therapy relative to other treatments in small-cell carcinoma of the lung. In: Thoracic Oncology. Choi NC, Grill0 H.C. eds. New York: Raven Press, 1983;233-256. 14 Cox JD, Azamia N, Byhart RW, Shin KH, Emami B, Pajak TF. A randomized phase I/II trial of hyperfractionated radiation therapy with total doses of 60.6 Gy to 79.2 Gy: possible survival benefit with > = 69.6 Gy in favorable patients with stage III non-small cell lung carcinoma: report of Radiation Therapy Oncology Group 83-l 1. J Clin Oncol 1990;8:1543-1555. 15 Coy P, Hodson BM, Payne D, Evans WK, Feld R, MacDonald AS, Osoba D, Pater JL. The effect of dose of thoracic irradiation on recurrence in patients with limited stage small cell lung cancer. Initial results of a Canadian Multicenter randomized trial. Int J Radiat Oncol Biol Phys. 1987;14:219-226. 16 Coy, P, Murray N, Pater J. et al. Optimizing the role of radiotherapy in the treatment of limited stage small cell lung cancer (abstract). Lung Cancer 1991; 7 (suppl):94. 17 Early Breast Cancer Trialists’ Collaborative Group. Treatment ofearly breast cancer. In: Worldwide evidence 1985-1990, Vol 1, Oxford Medical Publications, 1990. 18 Hodson DI, Malaker K, Meikle AL, Levitt M. Pilot studies of superfractionated radiotherapy and combination chemotherapy in limited oat cell carcinoma of the bronchus. Int J Radiat Oncol Biol Phys 1984;10:19411945. 19 Johnson BE, Grayson J, Woods E. et al. Limited stage small cell lung cancer treated with concurrent etoposide/ cisplatin plus bid chest radiotherapy. Proc Am Sot Oncol 1989;8:228. 20 Kies MS, Mira JC, Livingston RB. et al. Multimodal therapy for limited small cell lung cancer. A randomized study of induction combination chemotherapy with or without thoracic radiation in complete responders; and with widefield versus reduced volume radiation in partial responders. J Clin Oncol 1987;5:592-600. 21 Kwiatkowski DJ, Propert KJ, Carey RW, Choi N, Green M. A phase II trial of cyclophosphamide, etoposide and cisplatin with combined chest and brain radiotherapy in limited small-cell-lung cancer: a Cancer and Leukemia Group B study. J Clin Oncol 1987;5:1874-1879. 22 Le Chevalier T, Arriagada R, De The H. et al. Combination of chemotherapy and radiotherapy in limited small cell lung carcinoma: results of alternating schedule in 109 patients. NC1 Monogr 1988;6:335-338. 23 Looney WB, Hopkins HA. The integration of multifractionated radiotherapy into combined chemotherapeutic- radiotherapeutic approaches to lung cancer treatment. In: Arriagada R, Le Chevalier T, eds Treatment Modalities in Lung Cancer. Basel: Karger, Antibiot Chemother 1988;41:176183. 24 Looney WB. Special lecture: alternating chemotherapy and radiotherapy. NC1 Monogr 1988;6:85-94. 25 Mantyla M, Nuranen A. The treatment volume in radiation therapy of small cell lung cancer. IV World Conference of lung cancer. 1985, Toronto, Abstract 473:34. 26 McCracken JD, Janaki LM, Crowley JJ. et al. Concurrent chemotherapy/radiotherapy for limited small-cell lung carcinoma: A Soutwest Oncology Group Study. J Clin Oncol 1990; 8:892-898. 27 Mitchell JB, Morstyn G, Russo A., Carney DN. In vitro radiobiology of human lung cancer. Cancer Treat Symp 1985;2:3-10. 28 Murray N, Coldman A. Modelling the consequence of delay in the timing of thoracic irradiation in the combined modality treatment of limited stage small cell lung cancer (abstract). Lung Cancer 1991;7 (suppl): 152. 29 Murray N, Shah A, Brown E. et al. Alternating chemotherapy and thoracic radiotherapy with concurrent Cisplatin-Etoposide for limited-stage small cell carcinoma of the lung. Sem Oncol 1986;13:2430. 30 Nicolucci A, Grilli R, Alexanian A.A, Apolone G, Torri V, Liverati A. Quality, evolution, and clinical implications of randomized, controlled trials on the treatment of lung cancer. A lost opportunity for metaanalysis. JAMA, 1989;262:2101-2107. 31 Nielsen OS, Fode K, Bentzen SM, Schultz HP, Teenholdt S, Palshof T. Timing of radiotherapy and chemotherapy in limited stage small cell lung cancer. Final analysis (abstract). Eur J Cancer 1991; (suppl 2):Sl82. 32 Osterlind K, Hansen HH, Hansen HS, Dombernowsky P, Hansen M, Rorth M. Chemotherapy versus chemotherapy plus irradiation in limited small cell lung cancer. Results of a controlled trial with 5 years follow-up. Br J Cancer 1986;54:7-17. 33 Payne D, Arriagada R, Dombernowsky P. et al. The role of thoracic radiation therapy in small cell carcinoma of the lung: a consensus report. Lung Cancer 1989;5:135-138. 34 Perez CA, Bauer M, Edelstein S, Gillespie BW., Birch R. Impact of tumor control on survival in carcinoma of the lung treated with irradiation. Int J Radiat Oncol Biol Phys 1986;12:539-547.
263 35 Perez CA, Einhorn L, Oldham RK. et al. Randomized trial of radiotherapy to the thorax in limited small-cell carcinoma of the lung treated with multiagent chemotherapy and elective brain irradiation: a preliminary report. J Clin Oncol 1984;2:1200-1208. 36 Perez CA, Krauss S, Bartolucci AA. et al. Thoracic and elective brain irradiation with concomitant or delayed multiagent chemotherapy in the treatment of localized small cell carcinoma of the lung. A randomized prospective study by the Southeastern Cancer Study Group. Cancer 1981;47:2407-2413. 37 Perry MC, Baton WL, Propert KJ. et al. Chemotherapy with or without radiation therapy in limited small cell carcinoma of the lung. N Engl J Med 1987;316:912-918. 38 Popper KR. Conjectures and refutations. The Growth of Scientific Knowledge. 5th Ed. Routledge Publ. London, 1989. 39 Tubiana M, Arriagada R, Cosset J.M. Sequencing of drugs and radiation. The integrated alternating regimen. Cancer 1985;55:2131-2139. 40 Tubiana M. Repopulation in human tumors. A biological background for fractionation in radiotherapy. Acta Oncol 1988;27:83-88. 41 Turrisi A, Glover DJ. Thoracic radiotherapy variables: influence on local control in small cell lung cancer limited disease. Int J Radiat Oncol Biol Phys 1990;19:1473-1479. 42 Turrisi AT, Glover DJ, Mason BA. A preliminary report: concurrent twice-daily radiotherapy plus platinumetoposide chemotherapy for limited small cell lung cancer. Int J Radiat Oncol Biol Phys 1988;15:183-187. 43 White JE, Chen T, MacCracken J. et al. The influence of radiation therapy quality control on survival, response and sites of relapse in oat cell carcinoma of the lung. Preliminary report ofa Southwest Oncology Group Study. Cancer 1982;50:10841090. 44 Withers HR, Taylor JMG, Maciejewski B. The hazard of accelerated tumor clonogen repopulation during radiotherapy. Acta Oncol 1988;27:131-146.
253
reserved0169-5002BMO6.00
LUNG 00143
Controversies in the management of small cell lung cancer: thoracic radiotherapy in limited disease Rodrigo Department
Arriagada”,
of'Radiation Oncology,
Jean-Pierre
Pignonb and Thierry
Le Chevalier’
bMedical Statistics and “Medical Oncology, Institut Gustave- Roussy (IGR), Mllejuif, France
Summary The review of randomized trials shows that thoracic radiotherapy combined with chemotherapy decreases the risk of thoracic recurrences and provides a gain in long-term survival of approximately 6% in limited disease when compared to chemotherapy alone. Thoracic radiotherapy and combined radiochemotherapy approaches are composite parameters and the analysis of their effects can be equivocal. The purpose of this paper is to discuss such parameters, namely treatment toxicity, volumes to be treated, total tumor dose, fraction size, timing of radiochemotherapy approaches and the quality of radiotherapy. The analysis of these parameters permits a delineation of questions to be tested in future clinical trials.
Key words: Small cell lung cancer; Limited disease; Thoracic radiotherapy and chemotherapy
radiotherapy;
Combined
With each step forward, with each problem which we solve, we not only discover new and unsolved problems, but we also discover that where we believed that we were standing on firm and safe ground, all things are, in truth, insecure and in a state of flux
KARL R. POPPER Introduction
Thoracic radiotherapy (RT) was the main treatment of small cell lung cancer (SCLC) before the 1970s; long-term survival was poor (below 5% at 5 years). From then on, Correspondence to: Dr. R. Arriagada, Desmoulins, 94805 Villejuif, France.
Dept. of Radiation
Oncology, Institut Gustave-Roussy,
rue Camille
,rCdwaqloypt?J awn[oA paDnpaJ,aq~ u! 88~ 01 paJoduro3UJJB,AdeJaqlo!pwauxn[oAap!M, aql u! %ZE .wM awJ acwaJJnDaJ ~peJoq~~J~u![[t?JaAo aqL *[oz] IDafqnss!qluo papnpuo3 SI?M‘swayEd [6[ 8u!pn[3u! ‘[y-11 paz!uropueJ auo @IO *sa!pnis aA!yadsoJ~aJ 6q palJoddns aJo3aq lc[uo s!uogepuauu.uo~aJsfq~‘JaAaMoH $JEW aaJ3~13z-s*[e ~I!M ‘ICdoJaqlowaqD pauyap w aurn[oAJownl [eu$i!Jo aql $t?aJl 01 s![E[‘or]uogt2puauuuo9aJpJt?puels ayJ_ auqod loum$ pampa snsm
aurnIo.4 ~oumj ~a@~0
.Mo[aqpassnwp aq [[!MLayI put!suoyeJa)u! asay II!aua]Joduq [I?!~~JDJO aJt!sJalawk?Jed luatuloaJ1 Jaylo ‘JaAaMoH asanssyasoy uo IDa33a D!xolaA!pa[aue ~I!M pali?posst? aJe qa!q~ up!qnJoxop put!uph_uoa[qSE q3ns s%Jp3ouo!geJ~syupe aq)p!oncogJa3osaqp[noM $!'snql fawn[oA ayl u!qy~ pnwo[ sanssg aql u! Q.qEur pahlasqo uaaq aAeq asaqL *suogt?a pa,wpeJJ!. -![dwo~ihqua)oaJq)-a3![ uaAa pun aJaAasaDnpoJd uw ~a put!~330 suogewqwo3 walxa put2Ll!suqu!w30 t2ap! Jea[DIsaaaq 01 I[n9gyps!I!puti2 snoauai?oJajaq 1cJaAaJe dl!~!xol asoyhi 6%alEJlsynadt? uo SlJOdaJ‘~[a)wrn)Jo3u~ .asaas!p ~ut?u%y?wa$t?cNpt?Ja 01 s!UI!I? -Jay]hI0 U! JO13E3%U!@IU[aqj Uaaq ShM[" aAE?qSanSS!] [OUUOU UO Spa&a 3!XOJ 'h$!D!XOl [t?q)a[u! ast?aJDu! UB s!aJaql3! u!dynadeJaq~ ous!aJaq~leq$pu!wu! auJoq aq p[noqs11
Xdtrmyo!per aplr.~oqjJOasn aq) II! W~~AO.~UO~
yt2!~1 aJnln3u! palsa)aq 01 suoysanb aleaugap0) ldwa)leUB u! SJalaUJt?Jt?d asaq)[[I? ssn%!p 0) s!JadEd s!ql30asodJndaqL .dJaA![ap 6doJaqlo!peJ30 Qynbaql(i?)puvsaq~ooJdde[tyuanbas-uou SnsJaA[guanbas 0) f/idEJaqlo!pw ale[SnsJaAA[Jt?a (a)fansUO9DEJ3(p)taSOpJOUIn~["]Oj (3)fpa$EaJl aq 01 w.un[oA (q)tL~p!xol luauI)saJ) (B) :~uno~~eOIU!uayE1aq 01 aAt?qspadst? [t?$uaun?pun3 uaAag .[woA!nbaaqUKI s$Da33aJ!aql3o s!d[t?u~aq~ pun SJalaUIEJI?dal!sodUIo3 aJt?ddl?Jaq~OUIaqDO!pEJ paU!qUIODpUE La D!%?JOqL 'Sa[npaq%UO!~l?U!qrUOD JO3 ih!UU~ [wxydoue auyapola[qEun aJaM SfSd[i?Ut?-E)atU aqlu!slvpaqlq~fMapt?wsuosp~dwo~paJ -!pu!aqL .aseas!p pa~!uu[30lUaI.Uaii!EUt?w aql u! saqDeoJddeAdEJaq$OUIaqDO!pt?J pau!qwoD 30 asn [I?JaUa% aq$ slJoddns['I?A!AJns [[T?JaAo 30 sI.uJa~ uf lyauaq pa~t?J~suouIap OS aqL *[v] aDuaJa3uo3s!qlu! pa$uasaJds[t?p$ pazyopuw papuno3uoDun E[ 30 s!s@ut?-"law aA!suaqaJduIoD B dq paurryuo:, dpua3aJ SBM [vA!hlnsurral-8uo[ u! u@ sty1 *[EC‘6 ‘[I asEas!p pa~!uu[u! 8()-s I([alt?w!xoJdde3o[eA!Nns uua@o[u! u$!t! pas!uJoJdpu?i?sa~uaJ -Jn3aJD!DeJOql30?fS!J aq$ paWaJ3ap La +?JOq$ It?ql paMOqS S[l?!J) paZ!UIOpU&?J30 Ma!AaJ aqL .uo!ss!waJ ala[dwoDJaUe Jn33o 01 panuyuo3 sawaJJn3aJIsay ~0s uttq)aJ0t.u SE wa[qoJd ~0fa1.U t2pau!swaJ [o~)uo9[wo[ y3eJoql ‘auo[e~3 qI!M pa]eaJlsluagt?d30sapas lSaqaql uy uaAa 3vq3luap!AaaweDaq 1!‘JaAaMOH *(sJl?ad g $I?%8 6[alt?uqxoJddt?)[eA!"ns ruJa@uo[ u! uyG3 aleraporut!aAa!qDE01 a[qt? ~I?M~3 ‘auo[t? pasn uaqM .&x01 palt?[aJ -1uawwaJ) pawaJw_q 6[~ut?c~@&spua ‘kre3! ‘lyauaqt?paJaJ0 6[aJt2q ~a [EDO[leq)Bu! -uy[~ sa[npaqDs cynadEJaq$J!aq$u! asn s$!papJeDs!pswea$ ,GIE&~%uau@aJ uo!y?u!qwo:, U! _La 3!3l?JOq$ 30 UO!Sn[DU! aq$ JaAO kSJaAOJ$UODI?0) aS!JaA& LD qyM pap[a$ Salt?J aSUOdSaJq&q aqL g13S303uaur)EaJl aql-ro3 pasnIC[ap!Maqo~u&aq(~3)ddeJaq~otuaq3
255 arm; this difference was not significant. On the other hand, three retrospective studies including a total of 159 patients [25,36,43] showed that the use of inadequate portals could cause a two- to three-fold increase in intrathoracic recurrences. This significant adverse effect could be related to major protocol variations as the recurrence rates were over 50%. Our own data [8] showed no increase in the local relapse rate in the event of inadequate coverage - strictly defined - of the initial tumor: 36% versus 33% in the case of adequate coverage. This finding agrees with the results of the above-mentioned randomized trial, and it would suggest that tumor shrinkage after chemotherapy is sufficient to allow the tumor to be completely encompassed by ‘inadequate’ fields. We should not hastily conclude, however, that small fields should be recommended for the treatment of these patients. Retrospective series clearly indicate that very limited fields do not improve local control compared to treatments with chemotherapy alone. Three further arguments should be taken into account: (1) the difficulties encountered when attempting to define what adequate coverage means, (2) the relatively small number of patients included in these studies, and (3) the fact that the most frequent site of relapse was found inside the fields when the irradiation encompassed the original tumor volume [8]. This finding is possibly indicative of an insufficient dose level [15, 341. The amount of the surrounding lung which should be treated after chemotherapy is important inasmuch as it is closely linked to treatment toxicity and treatment failure. Further randomized and a priori well-defined studies, using strict definitions of original and reduced tumor volumes specified by the individual investigators, are necessary to settle this issue. Low versus moderate versus high total radiation dose The optimal radiation dose level is still unknown. It is worthwhile to point out that only one randomized trial has been conducted on this subject [15]. The dose levels compared were low versus moderate: 25 Gy in 10 fractions over 2 weeks and 37.5 Gy in 15 fractions over 3 weeks. The authors concluded that the use of a higher dose seemed to delay rather than prevent thoracic recurrence. All other data are only provided by retrospective studies; a summary is shown in Table 1. Clearly, doses below 40 Gy are unable to control local disease, but nothing can be stated about the effect of moderate (40-50 Gy) versus high dose (> = 60 Gy) radiotherapy. It is also interesting to observe the wide variation in the local
TABLE 1 Total tumor dose delivered by thoracic radiotherapy in SCLC and overall local failure” Tumor dose (Gy)
Local failure (96)
<40 40 4P 50 360
79-100 2343 16-39 25-42 4-25
“Modified from Turrisi [41] summarizing retrospective data from eight centers. bIncludes accelerated fractionation.
256 failure rate at each dose level. This is probably related to different definitions of local failure in each center: (i) recurrence after complete remission versus overall local failure; (ii) overall failure versus first site of failure; (iii) methodology used to report the results: total, censored or competing event approaches [5]. Our own data [5,8] did not show a significant difference in long-term local control in four consecutive alternating radiochemotherapy protocols which delivered a total radiation dose from 45 Gy up to 65 Gy. However, we should underline that this was also a retrospective analysis based on a limited number of patients and that radiotherapy was delivered in 3 split courses. There is certainly room for a randomized trial testing a moderate versus a high total radiation dose, for example 40 Gy versus 65 Gy or their biological equivalent in non-conventional fractionation. Conventional fractionation versus unconventional fractionation Conventional fractionation, defined as 5 fractions of 2 Gy per week, allows a total radiation dose as high as 70 Gy to be delivered in a limited mediastinal volume. The fraction size used in the 13 randomized trials included in the meta-analysis [4] varied between 2 and 4 Gy; a large fraction size could be a possible explanation for increased toxicity reported in some of these trials. The fraction size can be reduced and repeated a few hours later on the same day. This is called hyperfractionation, in which case, the treatment time remains the same but the total dose is slightly higher. The RTOG has conducted trials in non-SCLC using fractions of 1.2 Gy twice daily and increasing the total dose up to 79 Gy in a limited volume [14]. The rationale for the use of this type of schedule has been extensively developed in the relevant literature [40,44]. It is based on the premise that healthy normal tissue can be relatively spared while higher doses of RT can be delivered to tumor cells. This approach can be particularly interesting in ‘pure’ SCLC, as demonstrated by Mitchell et al. [27] with the survival curves obtained from cell cultures. Looney et al. [23] studied in the rat hepatoma 3924A the possibility of integrating multifractioned radiotherapy in an alternating RT-CT schedule and concluded in terms of a probable benefit for cure. Some clinical studies have been conducted in limited SCLC [12, 18, 19, 421. The results are still preliminary, but are encouraging. In Table 2 the series of Turrisi and Johnson are summarized, in which accelerated fractionation (1.5 Gy twice daily) was used allowing 45 Gy to be delivered in 3 weeks with the concurrent administration of etoposide and cisplatin. TABLE 2 Pilot studies on concurrent RT and CT in limited SCLC Author
n
CT
Dose RT MS (Gy/fraction) (weeks)
2-yr S (%)
McCracken (26) Murray (29) Kwiatkowski (21) Turrisi (42) Johnson (19)
156 IO 50 32 38
VCR-VP-CDDP CAVNP-CDDP CYC-VP-CDDP VP-CDDP VP-CDDP
45125 30/10 50125 45130” 45/3p
45 32 ? 51 65
MS: median survival. “Bifractionated radiotherapy.
13 78 63 _ 113
257 Our own series (protocol 010) is described in column 5 of Table 4. Patients were treated by an alternating RT-CT schedule in which the first course of RT was given with accelerated hyperfractionation using reduced fields without treating the spinal cord. This first course consisted of a dose of 21 Gy given in fractions of 1.4 Gy three times daily. The results were not different from those observed with our previous protocols, but we registered a higher, albeit nonsignificant, incidence of death unrelated to cancer and probably related to treatment toxicity [5]. Two patients died with sepsis or infectious disease, one with heart failure, one with renal failure, and one with lung fibrosis [A. There are two possible explanations for this toxicity: (i) a higher initial dose of cisplatin (150 mg/m’ given over 5 days); (ii) a 4-h time gap between the 3 daily fractions. Currently, a time gap equal to or above 6 h is recommended to permit the repair of sublethal lesions in normal tissues [44]. It should be noted that small changes may have an unexpected impact on toxicity, and this makes it more difficult to evaluate different combined schedules. For instance, for the first 4 patients in this series we used a dose of 25 Gy in the first multifractionated course. Three developed early and unusually severe esophagitis. By simply reducing the dose to 21 Gy (16% of the multifractionated course dose and 6% of the total radiation dose) the rate and intensity of the mucosa reaction reverted to previous levels. Accelerated multifractionated radiotherapy is an interesting approach in the management of SCLC. However, its integration in combined schedules should be carefully planned and evaluated in prospective studies exploring toxicity, as small changes in treatment parameters may have an unexpected impact on early and late toxicity. Early versus late toxicity There are at least two theoretical reasons why early thoracic RT should continue to be included. The first is the appearance of chemoresistant cells and the second is tumor repopulation during treatment. If CT and RT have to be given sequentially and in full doses, the consequence of this protracted administration may be a high level of tumor repopulation [39,44]. Occult distant metastases will continue to proliferate if the adminisTABLE 3 Alternating RT-CT protocols Protocol
002 (1980-81)
004 (1982-84)
006 (1985-86)
010 (1987-88)
Number of patients Cyclophosphamide (mp/m2) Doxorubicin d I (mg/m*) VP16213 (mg/m*) Methotrexate (mg/m*) Cisplatin (mg/m’) Thoracic RT (Gy) PC1 (GY)~ Maintenance CT (number of cycles)
28 300x4 40 75x3 400 _
81 300x4 40 75x3
64 300x3 40 100x3
29 300x4 40 75x5
100 20-20-I 5 30 8
120 20-20-25 24” 6
30x5 2l”-20-20 24 0
15-15-15 30 I2
“Multifractionated RT given by reduced radiation fields. bFractions of 3 Gy. “Random,ized for complete responders.
258 tration of CT is delayed. For example, if the tumor doubling time is 30 days, a delay of 2 months would permit a four-fold increase in the volume of metastases [39]. RT cannot be delayed for too long because CT often only has a limited effect on bulky tumors and can even elicit the development of radioresistance. In view of the uncertainty regarding the extent of repopulation, the shortest delay between CT and RT is advisable. More recently, Murray et al. [28] have developed a mathematical model along these lines. The indirect comparisons effectuated in the SCLC meta-analysis [4] did not provide a definite answer to whether a preference should be given to early or late radiotherapy. However, three randomized studies have directly tested this question and their results are summarized in Table 3. The NCIC trial showed a significant difference 0, = 0.016) in favor of early radiotherapy [16]. In the CALGB trial [37] there was a slight difference in favor of late radiotherapy which may be explained by the non-delivery of full chemotherapy doses in the early radiotherapy arm due to increased acute toxicity. This finding suggests that initial full doses of chemotherapy may have an influence on survival [3]. The Danish trial [31] did not show a significant difference. For the time being, we are of the opinion that the use of early radiotherapy should be favored, but further research is needed to justify this attitude. Sequential versus nonsequential approaches
We have previously analyzed theoretical and clinical reasons to prefer concurrent or alternating to sequential radiotherapy approaches [ 1,2,39]. The main advantage of concurrent combinations is that it is possible to deliver RT early and without intervals between the modalities. The NC1 trial [ 1l] reported a very high incidence of complications and lethal toxicity in spite of the survival benefit obtained with the combined treatment. For a while, these results were a reference to avoid concurrent administration. However, the high rate of complications may have been due to the use of a moderate concentrated radiation schedule (fractions of 2.66 Gy 5 times a week) and the administration of doxorubicin. These two factors were avoided in the two concurrent arms of the CALGB and the NCIC trials and the toxicity seemed acceptable. In addition, some reports of pilot studies with similar schedules have demonstrated the feasibility of this kind of combination [21, 26, 291 with interesting mid-term results as summarized in Table 2. The alternating approach has been defined as: ‘combined modality therapy in which radiotherapy is given on days of the chemotherapy cycle in which no chemotherapy is TABLE 4 Early versus late radiotherapy in limited SCLC. Results of two randomized trials Group
n
Initial day of RT
2-yr Survival (%I
CALGB [37]
125 145 155 153 97 98
d, dti d, d 84 d, d 126
24 30 40 33 20 18
NCIC [I61 Aarhus [31]
259
-
SURVIVAL
-
RFS
Fig. 1. Overall survival and recurrence-free survival in 202 patients with limited stage SCLC treated by the alternating RT-CT schedules at the lnstitut Gustave-Roussy and cooperating centers (19Wl988).
administered, without any delay in chemotherapy beyond that which would occur if chemotherapy were being given alone’ [l]. It was developed to reconcile the needs for early administration of both agents and for sequential delivery of each modality without a long interruption [39]. This alternating regimen is consistent with experimental data reported by Looney et al. [24] using their rat hepatoma model; many experiments combining radiotherapy and cyclophosphamide permitted the definition of an optimal schedule: alternating courses of CT with 3 courses of RT with an interval of 7 days between each treatment modality. This treatment obtained a cure rate of 60% compared to 0% when the animals were treated by a single modality. The clinical investigation was developed simultaneously but independently in our Institute. The use of 3 split courses of RT in these schedules is open to criticism because of a decreased tumor effect due to tumor repopulation between the rest periods. This is true for tumors responding poorly to CT, but in SCLC interdigitated CT probably precludes tumor repopulation between RT courses. The details of the therapeutic schedule have been published previously [6, 221. Thoracic RT was started one week after completion of the second CT course and was given for 12 days with one rest week between both modalities. Three courses of RT alternated with CT which was continued up to 6 courses in the induction phase were followed by a complete assessment including fiber-optic bronchoscopy to verify complete remission. The first pilot trial was started in May 1980 at the IGR and later these studies were extended to four cooperating centers. The summarized description of these trials, which included 202 patients, is shown in Table 4. The overall and disease-free survival curves are shown in Fig. 1. In summary, alternating RT-CT schedules are an interesting area of research worth developing in the coming years. Their reproducibility has been proven in different French centers and they have been included in prospective trials. It must be recalled that this
260 approach was also used in one of the randomized trials which demonstrated a significant advantage with the combined approach [35]. Even if concurrent or alternating approaches promise a therapeutic gain, no randomized trial has been published comparing these combinations to sequential ones. One ongoing EORTC trial is comparing an alternating to a sequential combined radiochemotherapy schedule; the direct comparison between concurrent and alternating radiochemotherapy combinations is being explored by a French GETCB study. A very large number of patients will probable be needed to detect a significant difference between these schedules. Quality of radiotherapy delivery All previous considerations are only valid if the quality of the radiotherapy delivered is good. Poor quality radiotherapy may overshadow benefits and increase toxicity. Modem megavoltage radiotherapy standards include: (a) an unequivocal definition of the volumes to be treated; (b) an unequivocal definition of the tumor dose; (c) prechemotherapy simulator film and CT-scan based planning; (d) optimal beam arrangement to cover the previous defined volume and to protect critical organs; (e) computer dosimetry to describe dose distribution; (l) check films in the treatment machine to ensure treatment reproducibility. Of course, these standards should be used in the randomized trials aiming to define optimal treatment.
Discussion Progress in improving survival in limited SCLC has been slow during the last two decades. The change in standard treatment from radiotherapy alone to chemotherapy alone in the 1970s increased the 5-year survival rate from roughly 5% to 10%. The combination of radiotherapy and chemotherapy in the 1980s gave an additional benefit of 5% (Fig. 1). These moderate effects on survival should be tested in randomized trials with a large number of patients. Small or medium-sized trials can result in equivocal answers or yield a mixture of ‘positive’ and ‘negative’ results; most of these contradictions can be explained by statistical variations. Comprehensive meta-analyses using individual data may provide a solution to this problem [ 171. A case in point is the study which evaluated the role of thoracic radiotherapy in limited SCLC [4]. One of the main merits of this study was to contradict the contention that meta-analysis is not a feasible means of evaluating the treatment effect in lung cancer [30]. It also showed the advantages of meta-analyses based on individual data rather than on a review of the literature. This above-mentioned study concluded that thoracic RT gives a survival benefit of 5 to 6% at 3 years; this is the mean effect of 13 randomized trials comprising more than 2,000 patients. The next step will be to determine the optimal combined radiochemotherapy schedule. One question was answered, but many others are now being posed as the combined approaches are multiparametric. It is inherent to scientific knowledge that each step forward raises new problems that are of an ever-increasing depth [38]. In this paper we have discussed some aspects of what optimal radiotherapy and timing of optimal radiochemotherapy could consist of. We have not discussed what would be the most appropriate chemotherapy: type of drugs, drug doses, administration schedule, etc.
261 It is interesting to note that we are looking for moderate effects. However, hundreds of patients are needed to test differences between two treatment attitudes. It would be useless to compare very similar treatments considering the number of patients required; years of research would be lost with this type of strategy. Treatments with rather extreme differences should be investigated (e.g. moderate versus high radiation doses, conventional versus unconventional fractionations, sequential versus non sequential schedules) in order to test the validity of our hypotheses. This type of trial should be initiated fairly rapidly. The first trial testing the role of thoracic RT in SCLC was started in 1976 [32]. It took 15 years to collect sufficient data worldwide to provide an unequivocal answer. It is therefore in the interest of our patients that we start - or continue - to investigate the new questions raised, so as to optimize the management of SCLC with combined modalities by conducting large randomized trials on a massive scale during the next few years.
Acknowledgements The authors wish to thank the Centre Hospitalier Intercommunal de Creteil, the Hopital A. B&cl&e and the Fondation Bergonie participants in the IGR SCLC protocols, Mrs. M. Tarayre for data collecting, Mrs. G. Feris for secretarial assistance and Ms. L. Saint Ange for her assistance in revising the manuscript. The IGR SCLC 010 protocol was partially supported by the research grants INSERMKNAMTS no. 883063 and ECC no. ST-A000309.
References 1 Arriagada R, Bertino JR, Bleehen NM, et al. Consensus report on combined radiotherapy and chemotherapy
5 6 7 8 9 10 II
modalities in lung cancer. In: Arriagada R, I_e Chevalier, T, eds. Treatment Modalities in lung cancer. Karger, Basel: Antibiot Chemother 1988;41:232-241. Arriagada R, Cosset JM, Le Chevalier T, Tubiana M. The value of adjunctive radiotherapy when chemotherapy is the major curative method. 1nt.J Radiat Oncol Biol Phys 1990;19:1279-1284. Arriagada R, De The H, Le Chevalier T, et al. Limited small cell lung cancer: a possible prognostic impact of initial chemotherapy doses. Bull Cancer 1989;76:605+615. Arriagada R, lhde DC, Johnson DH, et al. Meta-analysis of randomized trials evaluating the role of thoracic radiotherapy in limited small cell lung carcinoma (abstract). Lung Cancer 1991; 7 (suppl):98. Arriagada R, Kramar A, Le Chevalier T, De Cremoux H, for the French Cancer Centers’ Lung Group. Competing events determining relapse-free survival in limited small-cell lung carcinoma. J Clin Oncol 1992; Arriagada R, Le Chevalier T, Baldereyou P, et al. Alternating radiotherapy and chemotherapy schedules in small cell lung cancer, limited disease. Int J Radiat Oncol Biol Phys 1985;11:1461-1467. Arriagada R, Pellae-Cosset B, Baldeyrou P, et al. Initial high dose chemotherapy and multifractionated radiotherapy in limited small cell lung cancer (abstract). Lung Cancer 1991; 7 (suppl): 159. Arriagada R, Pellae-Cosset B, Cueto Ladron de Guevara J, et al. Alternating radiotherapy and chemotherapy schedules in limited small cell lung cancer: analysis of local chest recurrences. Radiother Oncol 1991;20:91-98. Arriagada R, Pignon JP, Le Chevalier T. Thoracic radiotherapy in small cell lung cancer: rationale for timing and fractionation. Lung Cancer 1989;5:237-247. Bleehen NM. Radiotherapy for small cell lung cancer. Chest 1986;89:268S-2768. Bunn PA Jr., Lichter AS, Makuch RW, et al. Chemotherapy alone or chemotherapy with chest radiation therapy in limited stage small cell lung cancer: a prospective randomized trial. Ann Intern Med 1987;106:655662.
262 12 Choi NC, Propert K, Carey R. et al. Accelerated radiotherapy followed by chemotherapy for locally recurrent small-cell carcinoma of the lung. A phase II study of Cancer and Leukemia Group B. Int J Radiat Oncol Biol Phys 13:263-266. 13 Choi NC. Reassessment of the role of radiation therapy relative to other treatments in small-cell carcinoma of the lung. In: Thoracic Oncology. Choi NC, Grill0 H.C. eds. New York: Raven Press, 1983;233-256. 14 Cox JD, Azamia N, Byhart RW, Shin KH, Emami B, Pajak TF. A randomized phase I/II trial of hyperfractionated radiation therapy with total doses of 60.6 Gy to 79.2 Gy: possible survival benefit with > = 69.6 Gy in favorable patients with stage III non-small cell lung carcinoma: report of Radiation Therapy Oncology Group 83-l 1. J Clin Oncol 1990;8:1543-1555. 15 Coy P, Hodson BM, Payne D, Evans WK, Feld R, MacDonald AS, Osoba D, Pater JL. The effect of dose of thoracic irradiation on recurrence in patients with limited stage small cell lung cancer. Initial results of a Canadian Multicenter randomized trial. Int J Radiat Oncol Biol Phys. 1987;14:219-226. 16 Coy, P, Murray N, Pater J. et al. Optimizing the role of radiotherapy in the treatment of limited stage small cell lung cancer (abstract). Lung Cancer 1991; 7 (suppl):94. 17 Early Breast Cancer Trialists’ Collaborative Group. Treatment ofearly breast cancer. In: Worldwide evidence 1985-1990, Vol 1, Oxford Medical Publications, 1990. 18 Hodson DI, Malaker K, Meikle AL, Levitt M. Pilot studies of superfractionated radiotherapy and combination chemotherapy in limited oat cell carcinoma of the bronchus. Int J Radiat Oncol Biol Phys 1984;10:19411945. 19 Johnson BE, Grayson J, Woods E. et al. Limited stage small cell lung cancer treated with concurrent etoposide/ cisplatin plus bid chest radiotherapy. Proc Am Sot Oncol 1989;8:228. 20 Kies MS, Mira JC, Livingston RB. et al. Multimodal therapy for limited small cell lung cancer. A randomized study of induction combination chemotherapy with or without thoracic radiation in complete responders; and with widefield versus reduced volume radiation in partial responders. J Clin Oncol 1987;5:592-600. 21 Kwiatkowski DJ, Propert KJ, Carey RW, Choi N, Green M. A phase II trial of cyclophosphamide, etoposide and cisplatin with combined chest and brain radiotherapy in limited small-cell-lung cancer: a Cancer and Leukemia Group B study. J Clin Oncol 1987;5:1874-1879. 22 Le Chevalier T, Arriagada R, De The H. et al. Combination of chemotherapy and radiotherapy in limited small cell lung carcinoma: results of alternating schedule in 109 patients. NC1 Monogr 1988;6:335-338. 23 Looney WB, Hopkins HA. The integration of multifractionated radiotherapy into combined chemotherapeutic- radiotherapeutic approaches to lung cancer treatment. In: Arriagada R, Le Chevalier T, eds Treatment Modalities in Lung Cancer. Basel: Karger, Antibiot Chemother 1988;41:176183. 24 Looney WB. Special lecture: alternating chemotherapy and radiotherapy. NC1 Monogr 1988;6:85-94. 25 Mantyla M, Nuranen A. The treatment volume in radiation therapy of small cell lung cancer. IV World Conference of lung cancer. 1985, Toronto, Abstract 473:34. 26 McCracken JD, Janaki LM, Crowley JJ. et al. Concurrent chemotherapy/radiotherapy for limited small-cell lung carcinoma: A Soutwest Oncology Group Study. J Clin Oncol 1990; 8:892-898. 27 Mitchell JB, Morstyn G, Russo A., Carney DN. In vitro radiobiology of human lung cancer. Cancer Treat Symp 1985;2:3-10. 28 Murray N, Coldman A. Modelling the consequence of delay in the timing of thoracic irradiation in the combined modality treatment of limited stage small cell lung cancer (abstract). Lung Cancer 1991;7 (suppl): 152. 29 Murray N, Shah A, Brown E. et al. Alternating chemotherapy and thoracic radiotherapy with concurrent Cisplatin-Etoposide for limited-stage small cell carcinoma of the lung. Sem Oncol 1986;13:2430. 30 Nicolucci A, Grilli R, Alexanian A.A, Apolone G, Torri V, Liverati A. Quality, evolution, and clinical implications of randomized, controlled trials on the treatment of lung cancer. A lost opportunity for metaanalysis. JAMA, 1989;262:2101-2107. 31 Nielsen OS, Fode K, Bentzen SM, Schultz HP, Teenholdt S, Palshof T. Timing of radiotherapy and chemotherapy in limited stage small cell lung cancer. Final analysis (abstract). Eur J Cancer 1991; (suppl 2):Sl82. 32 Osterlind K, Hansen HH, Hansen HS, Dombernowsky P, Hansen M, Rorth M. Chemotherapy versus chemotherapy plus irradiation in limited small cell lung cancer. Results of a controlled trial with 5 years follow-up. Br J Cancer 1986;54:7-17. 33 Payne D, Arriagada R, Dombernowsky P. et al. The role of thoracic radiation therapy in small cell carcinoma of the lung: a consensus report. Lung Cancer 1989;5:135-138. 34 Perez CA, Bauer M, Edelstein S, Gillespie BW., Birch R. Impact of tumor control on survival in carcinoma of the lung treated with irradiation. Int J Radiat Oncol Biol Phys 1986;12:539-547.
263 35 Perez CA, Einhorn L, Oldham RK. et al. Randomized trial of radiotherapy to the thorax in limited small-cell carcinoma of the lung treated with multiagent chemotherapy and elective brain irradiation: a preliminary report. J Clin Oncol 1984;2:1200-1208. 36 Perez CA, Krauss S, Bartolucci AA. et al. Thoracic and elective brain irradiation with concomitant or delayed multiagent chemotherapy in the treatment of localized small cell carcinoma of the lung. A randomized prospective study by the Southeastern Cancer Study Group. Cancer 1981;47:2407-2413. 37 Perry MC, Baton WL, Propert KJ. et al. Chemotherapy with or without radiation therapy in limited small cell carcinoma of the lung. N Engl J Med 1987;316:912-918. 38 Popper KR. Conjectures and refutations. The Growth of Scientific Knowledge. 5th Ed. Routledge Publ. London, 1989. 39 Tubiana M, Arriagada R, Cosset J.M. Sequencing of drugs and radiation. The integrated alternating regimen. Cancer 1985;55:2131-2139. 40 Tubiana M. Repopulation in human tumors. A biological background for fractionation in radiotherapy. Acta Oncol 1988;27:83-88. 41 Turrisi A, Glover DJ. Thoracic radiotherapy variables: influence on local control in small cell lung cancer limited disease. Int J Radiat Oncol Biol Phys 1990;19:1473-1479. 42 Turrisi AT, Glover DJ, Mason BA. A preliminary report: concurrent twice-daily radiotherapy plus platinumetoposide chemotherapy for limited small cell lung cancer. Int J Radiat Oncol Biol Phys 1988;15:183-187. 43 White JE, Chen T, MacCracken J. et al. The influence of radiation therapy quality control on survival, response and sites of relapse in oat cell carcinoma of the lung. Preliminary report ofa Southwest Oncology Group Study. Cancer 1982;50:10841090. 44 Withers HR, Taylor JMG, Maciejewski B. The hazard of accelerated tumor clonogen repopulation during radiotherapy. Acta Oncol 1988;27:131-146.