Influence of Interfraction Interval on Local Tumor Control in Patients With Limited-Disease Small-Cell Lung Cancer Treated With Radiochemotherapy

Int. J. Radiation Oncology Biol. Phys., Vol. 68, No. 2, pp. 426 – 432, 2007 Copyright © 2007 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/07/$–see front matter

doi:10.1016/j.ijrobp.2006.12.008

CLINICAL INVESTIGATION

Lung

INFLUENCE OF INTERFRACTION INTERVAL ON LOCAL TUMOR CONTROL IN PATIENTS WITH LIMITED-DISEASE SMALL-CELL LUNG CANCER TREATED WITH RADIOCHEMOTHERAPY BRANISLAV JEREMIC´ , M.D., PH.D.,

AND

BILJANA MILIC´ IC´ , M.D., M.SC.

Department of Oncology, University Hospital, Kragujevac, Serbia Purpose: To investigate the influence of interfraction interval (IFI) on local recurrence-free survival (LRFS) in patients with limited-disease small-cell lung cancer (LD SCLC) treated with accelerated hyperfractionated radiotherapy (Acc Hfx RT) and concurrent cisplatin and etoposide (PE). Methods and Materials: A total of 103 patients were treated with either “early” (Cycle 1) or “late” (Cycle 4) concurrent Acc Hfx RT/PE. Two daily fractions were nonrandomly given using an IFI of either 4.5–5.0 h (“shorter”) (n ⴝ 52) or 5.5– 6.0 h (“longer”) (n ⴝ 51). Results: The median LRFS and 5-year LRFS rate for all 103 patients were 52 months and 48%, respectively. Besides gender, Karnofsky performance status, and treatment group, IFI also influenced LRFS, whereas age and weight loss did not. When a multivariate model was used, IFI was marginally insignificant (p ⴝ 0.0770) as a predictor of LRFS. In terms of individual treatment groups, IFI was not significant in “early” Acc Hfx RT/PE but showed a strong trend in a “late” Acc Hfx RT/PE regimen. Although a shorter IFI led to a higher incidence of high-grade (>3) esophagitis, leukopenia, and infection, a correlation analysis of toxicities with all potential prognostic factors showed that a shorter IFI was not an independent predictor of any acute high-grade toxicity. Conclusion: “Shorter” IFI had a marginally insignificant influence on LRFS. A strong trend favoring it was observed in patients treated with “late” concurrent Acc Hfx RT/PE. This may be of interest because it could contribute to further understanding of potential biologic parameters influencing treatment outcome. © 2007 Elsevier Inc. Small-cell lung cancer, Radiotherapy, Chemotherapy, Interfraction interval, Accelerated hyperfractionation.

INTRODUCTION Radiotherapy (RT) has been widely introduced as a necessary part of the combined modality approach in patients with limited-disease small cell lung cancer (LD SCLC) owing to results of two meta-analyses published in the early 1990s (1, 2). With the widespread use of the cisplatin/ etoposide (PE) regimen and its low toxicity when combined with thoracic RT, concurrent thoracic RT and PE is now considered the standard treatment in LD SCLC. A recent meta-analysis (3) has confirmed the benefit of prophylactic cranial irradiation, at least in complete responders. Whereas earlier studies mostly used conventional or hypofractionated RT, recent studies used accelerated hyperfractionated (Acc Hfx) RT (4 –7). In some studies (5), hyperfractionation was showed to be superior to conventional fractionation, although in a Mayo clinic study (8) it was not so, likely as a consequence of a split introduced between the two Acc Hfx RT segments of the overall combined regimen.

In our continuous effort to optimize lung cancer treatment (both non–small cell and small cell) with Hfx RT and concurrent chemotherapy (CHT), we have used somewhat shorter (according to suggested radiobiological premises aiming to avoid excessive toxicity) interfraction intervals (IFI) between the two daily fractions (5, 9 –12). Our initial observation on the influence of IFI on outcome of patients with Stage III non-small-cell lung cancer (NSCLC) was that a shorter IFI (4.5–5.0 h) was superior to a longer one (5.5– 6.0 h) in terms of overall survival (13). Subsequently (14), the Radiation Therapy Oncology Group showed that the length of IFI did not influence either survival or incidence of acute Grade ⱖ3 esophagitis. Our recent analysis (15) of 301 patients treated with high-dose Hfx RT and concurrent CHT showed a very small advantage for shorter IFI, which prompted us to undertake an additional analysis of 536 patients in our database. These patients all had Stage III NSCLC treated with Hfx RT with or without concurrent CHT, and were enrolled during four prospective studies addressing an optimal treatment approach for this disease

Reprint requests to: Branislav Jeremic´, M.D., Ph.D., Applied Radiation Biology and Radiotherapy Section, International Atomic Energy Agency, P.O. Box 100, Wagramer Strasse 5, A-1400 Vienna, Austria. Tel: (⫹43) 1-2600-21666; Fax: (⫹43) 1-26007-

21668; E-mail: [email protected] Conflict of interest: none. Received Oct 27, 2006 and in revised form Dec 7, 2006. Accepted for publication Dec 7, 2006. 426

Influence of IFI on tumor control in LD SCLC

(16). “Shorter” IFI (4.5–5.0 h) led to better overall survival (p ⫽ 0.0000) and local recurrence-free survival (LRFS) (p ⫽ 0.0000). Multivariate analyses showed IFI to be an independent prognosticator of both overall survival and LRFS. These results were confirmed when we separated all patients into those treated with Hfx RT only and those treated with concurrent RT/CHT. Radiotherapy-related high-grade acute toxicity was not different between the two IFIs, but patients treated with a shorter IFI had a significantly higher incidence of hematologic toxicity (p ⫽ 0.002). No differences in late high-grade toxicities were found between the two interfraction intervals. Using regression analysis, it was shown that IFI was not a significant predictor of any acute or late high-grade (ⱖ3) toxicity. There are no such data for LD SCLC, although prevailing evidence suggests the use of Acc Hfx RT and concurrent PE in patients with LD SCLC (4 –7). The majority of studies supporting this approach used IFIs of ⱖ6 h (5– 8). However, we used IFIs of 4.5– 6.0 h in our study (4), similar to those used in NSCLC. The aim of this study was, therefore, to retrospectively investigate the influence of IFI duration on local control and toxicity in patients with LD SCLC treated with Acc Hfx RT and concurrent PE during a prospective, randomized, Phase III trial (4).

METHODS AND MATERIALS In the course of a prospective, randomized trial (4), patients with histologically or cytologically proven SCLC staged as LD (confined to one lung, the mediastinum, and the ipsilateral supraclavicular fossa), with no history of previous therapy, a Karnofsky performance score (KPS) of ⱖ50%, and adequate hematologic (leukocyte count ⱖ3000/␮L, platelet count ⱖ150.000/␮L), renal (serum creatinine concentration ⬍1.5 mg/dL), and hepatic (serum bilirubin level ⬍1.5 mg/dL) function were treated with Acc Hfx RT and CHT. Patients were excluded from the study if they were aged ⱖ70 years or had pleural effusion, serious cardiac or renal disease, or a history of any prior cancer in the preceding 5 years (except nonmelanoma skin cancer). Patients with superior vena cava syndrome were eligible if other criteria for LD were met. Patients were randomly assigned to either Acc Hfx RT and concurrent CHT consisting of 30 mg each of carboplatin (C) and etoposide (E) every RT day during the RT course (Weeks 1– 4), followed by four cycles of cisplatin/etoposide (PE) (P, 30 mg/m2 on Days 1–3 and E 120 mg/m2 on Days 1–3) given at 3-week intervals (Weeks 6, 9, 12, and 15) (Group 1), or to two cycles of PE (Weeks 1 and 4) followed by the same Acc Hfx RT and concurrent CE (Weeks 6 –9) and then by two additional cycles of PE (Weeks 11 and 14) (Group 2). Concurrent CE was administered during the intervals between the two daily fractions, 3 to 4 h after the first fraction (i.e., 1 to 2 h before the second fraction). At the end of Week 15, all patients with a complete response or partial response were given prophylactic cranial irradiation at Weeks 16 and 17. Accelerated Hfx RT was administered with 6 –10-MV photons from linear accelerators. Two daily fractions of 1.5 Gy were administered 5 times per week with an interfraction interval of 4.5– 6.0 h. After defining initial RT fields encompassing all gross disease with a 2-cm margin and the ipsilateral hilum plus the entire

●

B. JEREMIC´

AND

B. MILIC´ IC´

427

mediastinum with a 1-cm margin to give 36 Gy, a multifield technique was used to raise the total tumor dose to 54 Gy in 36 fractions in 18 treatment days over a period of 3.5 weeks. In Group 2, the field size was determined by the initial tumor volume. Patients were nonrandomly assigned to an IFI, and under no circumstance was a change from the “shorter” IFI (4.5–5.0 h) to a “longer” IFI (5.5– 6.0 h) permitted, although changes within an IFI (e.g., 4.5 h to 5.0 h or vice versa or from 5.5 h to 6.0 h or vice versa) were permitted. As in all other lung cancer hyperfractionation studies that have used these IFIs, this meant that the IFI group (“short” or “long”) to which patients were assigned remained unchanged throughout the course of Hfx RT. We always used an approach of nonrandomly assigning shorter or longer IFI, simply because no clear data on its impact existed at the time we designed this study. Consequently, no stratification was justified, and we simply tried to “mimic” stratification with this, rather unusual, approach. No dose modifications and no treatment delays for either RT or CHT during the treatment were allowed. Differences in patient characteristics and the incidence of toxicity were evaluated by the Pearson chi-square test. Curves for local control were generated using the Kaplan-Meier method, and the differences between pairs of curves were examined by the log–rank method. Univariate and multivariate Cox regression models were used to test the independent influence of various potential prognostic factors. Using the Cox proportional hazards model, the interaction of each prognostic factor and its effect on local control were analyzed, and baseline local control curves were generated for each IFI group. Despite the fact that local recurrence is sometimes hard to assess and that overall survival may have been an endpoint, local control was used as an endpoint because it is expected that IFI should have influenced local control first and only then overall survival.

RESULTS This analysis was based on the data collected after the closure of the original study for the initial report (4), and no update was subsequently done. Of the 103 patients available for the original analysis (4), there were 18 of 52 patients (35%) alive with no evidence of the disease (NED) in Group 1, whereas in Group 2, there were 8 of 51 NED patients. Median follow-up for all 103 patients was 31 months (range, 2–77 months); whereas for the 26 NED patients it was 57.5 months (range, 45–74 months). Pretreatment and treatment characteristics of the 103 patients are shown in Table 1. The median LRFS for all 103 patients was 52 months, and the 1-, 2-, 3-, 4-, and 5-year LRFS rates were 84%, 78%, 64%, 55%, and 48%, respectively (Fig. 1). Local recurrence-free survival analysis according to various potential prognostic factors showed that women achieved significantly better LRFS than men (Table 2). Age did not impact LRFS, and there was no difference between patients aged ⬍60 years and those aged ⱖ60 years (p ⫽ 0.2951). Patients having a KPS of 90 –100 had significantly better LRFS than those having a KPS of 50 – 80. Pretreatment weight loss of ⬎5% did not influence LRFS (p ⫽ 0.9765). Treatment group influenced LRFS in a way that favored

428

I. J. Radiation Oncology

●

Biology

●

Physics

Table 1. Pretreatment and treatment characteristics of patients according to interfraction interval Characteristic Gender Male Female Age (y) ⬍60 ⱖ60 KPS 50–80 90–100 Weight loss ⱕ5% ⬎5% Group 1 2

4.5–5.0 h (n)

5.5–6.0 h (n)

29 23

33 18

20 32

33 18

18 34

34 17

21 31

33 18

26 26

26 25

p 0.354 0.008 0.001 0.013 0.921

Abbreviation: KPS ⫽ Karnofsky performance status.

patients treated with “early” concurrent Acc Hfx RT/CE and PE (p ⫽ 0.0064). Finally, patients treated with a “shorter” IFI had significantly better LRFS than those treated with a “longer” IFI (p ⫽ 0.0105) (Fig. 2). Univariate analysis with a Cox model confirmed this observation, showing that age and weight loss did not influence LRFS. However, when a multivariate model was used, it was shown that the strongest predictor of LRFS was treatment group, followed by KPS and gender, whereas IFI was marginally insignificant (p ⫽ 0.077) as a predictor of LRFS. When a multivariate analysis was performed according to treatment group, in Group 1 IFI was not taken into account for this analysis because none of the examined factors except KPS seemed to influence LRFS in the univariate Cox analysis. In Group 2, IFI influenced LRFS in both Kaplan-Meier survival analysis and univariate Cox analysis, but it lost its significance when a multivariate Cox model was used (Table 3). Using the Cox proportional hazards model to analyze the interaction of each prognostic factor (excluding age and weight loss, which were not shown to be independent prognosticators of LRFS) and its effect upon LRFS, baseline local control curves were generated for each of the IFI groups. The median times to an event were 44 months and 33 months for the shorter and longer intervals, respectively, whereas 1-year, 2-year, and 3-year LRFS rates were 95%, 92%, and 57% vs. 94%, 88%, and 27%, respectively (Fig. 3). Although the difference was 30% at 3 years, the difference was not significant owing to the small (1– 4%) difference at 1–2 years. Acute high-grade (ⱖ3) toxicity is shown in Table 4. There was no difference in the incidence of alopecia, anemia, thrombocytopenia, and nausea and vomiting between the two IFI groups. Patients treated with a shorter IFI had significantly more frequent esophagitis, infection, and leukopenia. Because of their low incidence in this group of patients, late high-grade bronchopulmonary (n ⫽ 1) and esophageal (n ⫽ 2) toxicity were not evaluated.

Volume 68, Number 2, 2007

To evaluate which potential prognostic factor could have predicted acute high-grade toxicity, regression analysis was performed next using the Cox proportional hazards model (Table 5). None of the variables influenced the occurrence of alopecia, nausea and vomiting, anemia, thrombocytopenia, and bronchopulmonary toxicity. Karnofsky performance status remained the sole prognosticator of acute high-grade leukopenia and infection. Interfraction interval influenced leukopenia, infection, and esophagitis in a univariate model, but it lost significance when multivariate modeling was performed for leukopenia and infection. Interfraction interval was the only variable that influenced the occurrence of esophagitis, but only in univariate analysis, which prevented us from performing a multivariate regression analysis using esophagitis as an endpoint.

DISCUSSION Reports of Hfx RT regimens in a number of tumor sites have mainly involved an IFI of 4 – 8 h (17–20). These IFIs were deduced from the estimated repair half-times in normal tissues (21, 22). With a short IFI, late toxicity in patients with head-and-neck cancer (23, 24) was increased, seemingly in agreement with the analysis (25) and indicating repair half-times as long as 3.8 – 4.9 h in three latereacting tissues. These data indicate that recommended IFIs of up to 8 h are not sufficient for full repair of early- and late-reacting tissues, leading to an increase in toxicity. Whereas studies of IFI in head-and-neck cancer (23) have confirmed this observation, studies of NSCLC have mixed results, indicating both that IFI does not influence overall treatment outcome (14, 15) and that it does (13, 16), favoring short IFI. Interfraction interval also did not influence treatment-related toxicity, except hematologic, the latter being attributed to the use of concurrent CHT. It is reiterated that there may be a difference in the repair kinetics between early-reacting normal tissues (more rapidly repairing dam-

Fig. 1. Local recurrence-free survival for all patients.

Influence of IFI on tumor control in LD SCLC

●

B. JEREMIC´

AND

B. MILIC´ IC´

429

Table 2. Local recurrence-free survival according to various potential prognostic factors LRFS (%) Factor Gender Male Female Age (y) ⬍60 ⱖ60 KPS 50–80 90–100 Weight loss ⱕ5% ⬎5% IFI (h) 4.5–5.0 5.5–6.0 Group 1 2

MLRFST (mo)

1y

2y

3y

4y

5y

47 NA

77 95

68 93

56 76

49 64

37 64

45 NA

81 88

73 84

61 67

49 61

44 51

25 NA

68 100

51 100

0 86

0 73

0 64

51 52

90 79

80 77

59 71

52 58

48 52

NA 41

90 78

86 70

74 52

68 35

58 35

NA 44

96 74

89 67

73 55

66 44

61 33

p

Univariate p

Multivariate p

0.0195

0.0239

0.0327

0.2951

0.3005

—

0.0000

0.0000

0.0060

0.9765

0.9766

—

0.0105

0.0131

0.0770

0.0064

0.0087

0.0004

Abbreviations: MLRFST ⫽ median local recurrence-free survival time; LRFS ⫽ local recurrence-free survival; NA ⫽ not achieved yet; KPS ⫽ Karnofsky performance status; IFI ⫽ interfraction interval.

age) and tumors (less rapidly repairing the damage) and late-reacting normal tissues, spared by hyperfractionation. All of these issues are virtually unknown in SCLC, although several important studies have used Hfx RT regimens. As a rule, all of these studies used 1.5 Gy b.i.d. fractionation (4 –7), following initial promising results of a Phase II study (26), and most North American hyperfractionation treatment schedules mandate an IFI of ⬎6 h. Our IFIs were 4.5– 6.0 h, assuming the radiobiologically though not clinically tested (at that time) differential repair between tumor and normal tissues expected to occur with hyperfractionation. In the prospective, randomized study used as a

Fig. 2. Local recurrence-free survival according to interfraction interval: “short” (4.5–5.0 h) (solid line) and “long” (5.5– 6.0 h) (dashed line).

source for this analysis, we delivered a total of 54 Gy using 1.5 Gy b.i.d. fractionation with concurrent low-dose daily CE, followed by four cycles of PE chemotherapy in patients with LD SCLC. Patients treated with a “shorter” IFI achieved significantly better LRFS than those treated with a “longer” IFI (p ⫽ 0.0105). However, when a multivariate model was used, the influence of IFI became marginally insignificant (p ⫽ 0.077) as a predictor of LRFS. The timing of administration of combined RT and CHT is an important issue in the treatment of LD SCLC. Recent meta-analyses (27–29) showed diverging findings, and a subsequent editorial (30) summarized the findings and highlighted advantages of early administration of concurrent RT/CHT over the late RT/CHT in LD SCLC. Our original prospective trial design (4) gave us an opportunity to investigate the influence of IFI on LRFS in both early and late RT/CHT groups. In Group 1 (early concurrent RT/CHT) IFI did not influence LRFS in both Kaplan-Meier analysis and univariate Cox analysis and was not entered into a multivariate model. In Group 2, IFI influenced LRFS in both Kaplan-Meier survival analysis and univariate Cox analysis but was insignificant again in the multivariate analysis, showing only a strong trend (p ⫽ 0.077). Because early concurrent RT/CHT was a highly significant and independent variable, it would be interesting to speculate about possible disassociation of findings between early and late concurrent RT/CHT from the standpoint of accelerated repopulation of tumor clonogens. It may be possible that in the early group, in which intensive treatment was advantageous, short IFI did not influence LRFS owing to the effect of early “timing”. Contrary to that, in the late group, with significantly inferior local tumor control than that observed

430

I. J. Radiation Oncology

●

Biology

●

Physics

Volume 68, Number 2, 2007

Table 3. Effect of interfraction interval on median time to local recurrence (months) of patients with LD SCLC MLRFST (mo) Group (n)

IFI 4.5–5.0 h

IFI 5.5–6.0 h

Log–rank p

Univariate p

Multivariate p*

All patients (103) Group 1 (52) Group 2 (51)

NA NA 52

41 NA 28

0.0105 0.0043 0.0189

0.0131 0.2397 0.0256

0.0770 — 0.0938

Abbreviations: MLRFST ⫽ median local recurrence-free survival time; IFI ⫽ interfraction interval; LD SCLC ⫽ limited-disease small-cell lung cancer; NA ⫽ not achieved. * When corrected (see Table 2 for all variables entered into the model).

in the early group, treatment started with CHT, which may have adversely influenced local control, as seen from the results of the original study (4). There, with more tumor clonogens being present at the time of the commencement of Acc Hfx RT, shorter IFI may have led to more cell kill than longer IFI, materialized in borderline insignificant influence of shorter IFI on LRFS in that group of patients. Baseline local control curves and the median times to an event were 44 months and 33 months for the shorter and longer intervals, respectively, whereas 1-, 2-, and 3-year LRFS rates were 95%, 92%, and 57% vs. 94%, 88%, and 27%, respectively (Fig. 3). The difference was not significant owing to the low (1– 4%) difference at 1–2 years, although it rose to 30% at 3 years. Because of the somewhat small patient numbers in the “late” concurrent RT/CHT group, we did not construct baseline local control curves for patients treated with a longer IFI in that group. There was no difference in the incidence of alopecia, anemia, thrombocytopenia, and nausea and vomiting between the two IFI groups, but patients treated with shorter IFI had significantly more frequent esophagitis, infection, and leukopenia. When we evaluated which potential prognostic factors could have been predictive for various kinds

of acute high-grade toxicity, regression analysis showed that IFI did not independently predict any of these highgrade toxicities. The small number of toxicities may suggest a need for caution in interpretation of the data, but these results may also fit the long-standing belief that effects on normal tissues (especially those expressing fast response) should mimic those observed in tumors. Although esophageal toxicity was shown to be the principal concern in the majority of Acc Hfx RT regimens when given concurrently with high-dose CHT, our original study (4) showed that even a somewhat higher total dose (54 Gy) using the same fractionation regimen (1.5 Gy) can be of lower toxicity with the use of concurrent low-dose daily CE (even if short IFI were administered) compared with toxicities observed with similar Acc Hfx RT regimens given concurrently with high-dose CHT (5). In addition, although 5-year survival rates seem similar between our Group 1 (early Acc Hfx RT) and the Hfx RT group of an Intergroup study (5), in our initial study we achieved a median survival time of 30 months (vs. 23 months in the Intergroup study). This may indicate that higher total RT dose (54 Gy vs. 45 Gy, both using 1.5 Gy b.i.d.) may have a

Table 4. High-grade (ⱖ3) toxicity according to interfraction interval Interfraction interval Toxicity

Fig. 3. Baseline local control curves taking into account three independent prognosticators (gender, Karnofsky performance status, and treatment group) of local control. Solid and dashed lines as in Fig. 2.

Alopecia Yes No Anemia Yes No Esophagitis Yes No Thrombocytopenia Yes No Leukopenia Yes No Infection Yes No

4.5–5.0 h

5.5–6.0 h

2 50

6 45

5 47

6 45

20 32

8 43

18 34

13 38

25 27

13 38

11 41

3 48

p 0.133 0.724 0.009 0.313 0.018 0.024

Influence of IFI on tumor control in LD SCLC

●

B. JEREMIC´

AND

B. MILIC´ IC´

431

Table 5. Correlation of various pretreatment and treatment variables with acute high-grade (ⱖ3) toxicity Variable

Alopecia

Nausea and vomiting

Anemia

Platelets

WBC

Infection

Esophageal

Bronchopulmonary

Gender Age KPS Weight loss IFI Group

0.175 0.520 0.449 0.888 0.136 0.449

0.086 0.259 0.753 0.234 0.314 0.753

0.807 0.397 0.727 0.628 0.727 0.361

0.146 0.407 0.317 0.915 0.317 0.063

0.235 0.530 0.000/0.000* 0.111 0.017/0.234* 0.377

0.157 0.066 0.000/0.003* 0.181 0.024/0.205* 0.969

0.608 0.291 0.168 0.888 0.009 0.705

0.419 0.306 0.324 0.296 0.324 0.324

Abbreviations: WBC ⫽ white blood cell count; KPS ⫽ Karnofsky performance status; IFI ⫽ interfraction interval. * Multivariate; all other p values univariate.

potential for an improvement in overall survival, especially in the light of different IFIs used in these two studies. As with all retrospective studies, this study carries hidden biases that necessitate careful interpretation of its results. Besides the nonrandom nature of the IFI assignment, it seems that there was a tendency of involved radiation oncologists to assign shorter IFI to patients with higher KPS and a longer IFI to those with lower KPS. Age seems to have had a similar influence in the decision-making process, younger patients being more frequently treated with a shorter IFI. This may have impacted on overall results in this study, although age alone was not shown to be an independent prognosticator of LPFS, whereas KPS and treatment group did influence it. Gender also favored women, a finding that has been frequently observed in a number of lung cancer studies, although that was usually using overall survival as an endpoint. No rational explanation currently exists for a “biologic” difference between men and women that could explain better local control in women. Indeed, in the original study (4), although female gender carried an improvement in univariate analysis, in the multivariate analysis it did not independently influence survival.

In conclusion, this is the first study to explore the influence of IFI on LRFS in patients with LD SCLC treated with Acc Hfx RT/CHT. Although overall results showed no influence of a shorter IFI on LRFS, a strong trend favoring it was observed in the late Acc Hfx RT (i.e., when treatment started with CHT). Our hypothesis is that, in this case, the shorter IFI may have compensated for poorer local control, observed as a consequence of the initial use of CHT. Alternatively, results of our study may imply that such an effect does not exist at all, or at least not when early concurrent RT/CHT is administered. Also, it may be entirely possible that with this short range of IFIs the difference, if it exists, could not be detected, therefore requiring different comparisons (e.g., 5 h vs. 8 h). The results of this rather small and single-institution study should be confirmed in a prospective, randomized trial comparing shorter vs. longer IFI in patients with LD SCLC, starting their combined modality treatment with CHT. We believe the issue of optimal IFI remains of interest because it could contribute to further understanding of potential biologic parameters, and by manipulating IFI, influence treatment outcome.

REFERENCES 1. Pignon JP, Arriagada R, Ihde DC, et al. A meta-analysis of thoracic radiotherapy for small-cell lung cancer. N Engl J Med 1992;327:1618 –1627. 2. Warde P, Payne D. Does thoracic radiation improve survival and local control in limited-stage small cell carcinoma of the lung? J Clin Oncol 1992;10:890 – 895. 3. Auperin A, Arriagada R, Pignon JP, et al. Prophylactic cranial irradiation for patients with small-cell lung cancer in complete remission. Prophylactic Cranial Irradiation Overview Collaborative Group. N Engl J Med 1999;341:476 – 484. 4. Jeremic B, Shibamoto Y, Acimovic L, et al. Initial versus delayed accelerated hyperfractionated radiation therapy and concurrent chemotherapy in limited small cell lung cancer. J Clin Oncol 1997;15:893–900. 5. Turrisi AT, Kim K, Blum R, et al. Twice-daily compared with once-daily thoracic radiotherapy in limited small-cell lung cancer treated concurrently with cisplatin and etoposide. N Engl J Med 1999;340:264 –271. 6. Skarlos DV, Samantas E, Briassoulis E, et al. Randomized comparison of early versus late hyperfractionated thoracic irradiation concurrently with chemotherapy in limited disease small-cell lung cancer: A randomized phase II study of the

7.

8.

9.

10.

11.

Hellenic Cooperative Oncology group (HeCOG). Ann Oncol 2001;12:1231–1238. Takada M, Fukuoka M, Kawahara M, et al. Phase III study of concurrent versus sequential thoracic radiotherapy in combination with cisplatin and etoposide for limited-stage small-cell lung cancer: Results of the Japan Clinical Oncology Group study 9104. J Clin Oncol 2002;20:3054 –3060. Bonner JA, Sloan JA, Shanahan TG, et al. Phase III comparison of twice-daily split-course irradiation versus once-daily irradiation for patients with limited stage small-cell lung carcinoma. J Clin Oncol 1999;17:2681–2688. Jeremic B, Shibamoto Y, Acimovic L, et al. Randomized trial of hyperfractionated radiation therapy with or without concurrent chemotherapy for stage III non-small-cell lung cancer. J Clin Oncol 1995;13:452– 458. Jeremic B, Shibamoto Y, Acimovic L, et al. Hyperfractionated radiation therapy with or without concurrent low-dose daily carboplatin/etoposide for stage III non-small-cell lung cancer: A randomized study. J Clin Oncol 1996;14:1065– 1070. Jeremic B, Shibamoto Y, Milicic B, et al. Concurrent radiochemotherapy for patients with stage III non-small cell lung

432

12.

13.

14.

15.

16.

17.

18.

19.

20.

I. J. Radiation Oncology

●

Biology

●

Physics

cancer (NSCLC). Long-term results of a phase II study. Int J Radiat Oncol Biol Phys 1998;42:1091–1096. Jeremic B, Shibamoto Y, Acimovic LJ, et al. Hyperfractionated radiation therapy and concurrent low-dose, daily carboplatin/ etoposide with or without week-end carboplatin/etoposide chemotherapy in stage III non-small-cell lung cancer: A randomized trial. Int J Radiat Oncol Biol Phys 2001;50:19 –25. Jeremic B, Shibamoto Y. Effect of interfraction interval in hyperfractionated radiotherapy with or without concurrent chemotherapy for stage III nonsmall cell lung cancer. Int J Radiat Oncol Biol Phys 1996;34:303–308. Werner-Wasik M, Scott C, Graham ML, et al. Interfraction interval does not affect survival of patients with non-small cell lung cancer treated with chemotherapy and/or hyperfractionated radiotherapy: A multivariate analysis of 1076 RTOG patients. Int J Radiat Oncol Biol Phys 1999;44:327–331. Shibamoto Y, Jeremic B, Acimovic LJ, et al. Influence of interfraction interval on the efficacy and toxicity of hyperfractionated radiotherapy in combination with concurrent daily chemotherapy in stage III nonsmall cell lung cancer. Int J Radiat Oncol Biol Phys 2001;50:295–300. Jeremic B, Milicic M, Dagovic A, et al. Interfraction interval in patients with stage III non-small cell lung cancer treated with hyperfractionated radiation therapy with or without concurrent chemotherapy: Final results in 536 patients. Am J Clin Oncol 2004;27: 616 – 625. Brizel DM, Albers ME, Fisher SR, et al. Hyperfractionated irradiation with or without concurrent chemotherapy for locally advanced head and neck cancer. New Engl J Med 1998; 338:1798 –1804. Jeremic B, Grujicic D, Antunovic V, et al. Hyperfractionated radiation therapy (HFX RT) followed by multiagent chemotherapy (CHT) in patients with malignant glioma: A phase II study. Int J Radiat Oncol Biol Phys 1994;30:1179 –1185. Jeremic B, Shibamoto Y, Milicic B, et al. Hyperfractionated radiation therapy with or without concurrent low-dose cisplatin in locally advanced squamous cell carcinoma of the head and neck: A prospective randomized trial. J Clin Oncol 2000; 18:1458 –1464. Fu KK, Pajak TF, Trotti A, et al. A Radiation Therapy Oncology Group (RTOG) phase III randomized study to compare hyperfractionation and two variants of accelerated frac-

Volume 68, Number 2, 2007

21. 22.

23.

24.

25.

26.

27.

28.

29.

30.

tionation to standard fractionation radiotherapy for head and neck squamous cell carcinomas: First report of RTOG 9003. Int J Radiat Oncol Biol Phys 2000;48:7–16. Rojas A, Joiner MC. The influence of dose per fraction on repair kinetics. Radiother Oncol 1989;14:329 –336. Thames HD Jr, Withers HR, Peters JJ. Tissue repair capacity and repair kinetics deduced from multifractionated or continuous irradiation regimens with incomplete repair. Br J Cancer 1984;49(Suppl. VI):263–269. Cox JD, Pajak TF, Marcial VA, et al. ASTRO plenary: Interfraction interval is a major determinant of late effects, with hyperfractionated radiation therapy of carcinomas of upper respiratory and digestive tracts: Results from Radiation Therapy Oncology Group protocol 8313. Int J Radiat Oncol Biol Phys 1991;20:1191–1195. Fu KK, Pajak TF, Marcial VA, et al. Late effects of hyperfractionated radiotherapy for advanced head and neck cancer: Long-term follow-up results of RTOG 83–13. Int J Radiat Oncol Biol Phys 1995;32:577–588. Bentzen SM, Saunders MI, Dische S. Repair halftimes estimated from observations of treatment-related morbidity after CHART or conventional radiotherapy in head and neck cancer. Radiother Oncol 1999;53:219 –226. Turrisi AT 3rd, Glover DJ, Mason BA. A preliminary report: Concurrent twice-daily radiotherapy plus platinum-etoposide chemotherapy for limited small cell lung cancer. Int J Radiat Oncol Biol Phys 1988;15:183–187. Fried DB, Morris DE, Poole C, et al. Systematic review evaluating the timing of thoracic radiation therapy in combined modality therapy for limited-stage small-cell lung cancer. J Clin Oncol 2004;22:4785– 4793. Huncharek M, McGarry R. A meta-analysis of the timing of chest irradiation in the combined modality treatment of limited-stage small cell lung cancer. Oncologist 2004;9:665– 672. Pijls-Johannesma MCG, De Ruysscher D, Lambin P, et al. Early versus late chest radiotherapy for limited stage small cell lung cancer. Cochrane Database Syst Rev 2005;(1):CD004700. Jeremic´ B. Timing of concurrent radiotherapy and chemotherapy in limited-disease small-cell lung cancer: “Meta-analysis of meta-analyses.” Int J Radiat Oncol Biol Phys 2006;64:981– 982.