• Three-week schedule

After the protocol was amended, a total of 119 three-week-scheduled treatment cycles were administered to 40 pts. Thirty-nine pts were evaluable for treatment compliance and 110 cycles were fully evaluable for toxicity.
Grade 3/4 thrombocytopenia was registered in 2 (2%) cycles/2 pts and in 2 (2%) cycles/1 pt, respectively; it was not associated with bleeding events, with recovery within a week.
Transient G3/4 neutropenia occurred in 3 (3%) cycles/3 pts and in 1 (1%) cycle, respectively; overall, 2 episodes of febrile neutropenia were observed.
Non-hematological toxicity consisted of G3 vomiting in 1 cycle. Moreover, 1 pt experienced a transient GCB-induced systemic capillary-leak syndrome, with complete symptom recovery within 2 days after intravenous administration of Furosemide and high-dose Dexamethasone.
As reported with the four-week schedule, flu-like symptoms were occasionally seen and never required treatment modifications.
Overall, treatment-related toxicity caused a dose-reduction or treatment omission in 10 cycles (10%)/7 pts (17%); no patients required chemotherapy discontinuation. Reasons for dose reduction - treatment omission are listed in table 3.

Table 3. Three-week schedule: reasons for dose-modifications/treatment omissions

• Radiation therapy

Acute RT toxicity (RTOG) was mainly related to esophageal mucositis (grade 2 in 25 patients, grade 3 G3 in 4 patients). Concerning late toxicity, no grade 3 esophagitis occurred; on the opposite, one grade-3 lung toxicity was diagnosed

• Dose-Intensity

Dose intensity has been calculated for each schedule of administration. With the schedule CDDP 100 mg/m² dd 1, 28 and GCB 1000 mg/m² dd 1, 8, 15, 28, taking into account dose reduction for each drug as well as any delay in drug administration, the actual delivered dose intensities of both CDDP and GCB, calculated to the start dose, were 86.5% of the intended dose level, at all dose levels. Namely it was 90% for CDDP and 82.7% for GCB. Cisplatin was administered within the 67% to 100% range of the intended dose in 100% of the courses. As a result of omissions and dose reduction 9 out of the 10 pts treated with this schedule did not received the intended dose of GCB. Range of administration were included between 70% and 91.6% of the intended dose. With the schedule CDDP 80 mg/m² dd 1, 21 and GCB 1250 mg/m² dd 1, 8, 21, the actual delivered dose intensities of both drugs, calculated to the start dose, were 99.55% of the intended dose level. Namely it was 99.7% for CDDP and 99.5% for GCB. Cisplatin and GCB was administered within the 91% to 100% and within 93.3% and 100% range of the intended dose in 100% of the courses, respectively. As a result of omissions and dose reductions 10 of the 110 evaluable courses were not administered at the intended dose of GCB.

Response to induction therapy

All patients underwent tumor-response assessment after 2 and 3 cycles of treatment. All patients were evaluable for response; all the CT scans of responding patients were reviewed. Responses to chemotherapy are listed in table 4.

Table 4. Antitumor activity

Thirty-seven out of 50 patients (74%, 95% CI: 60-85%) achieved a partial response, 7 had stable disease and 6 had disease progression. One nodal pathological complete remission was obtained among the 24 pts that underwent surgery.
Activity was seen with both the three-week and four-week schedules, with 6 out of 10 and 31 out of 40 partial responses and 1 and 5 tumor progressions, respectively.
All the 6 pts with progressive disease were staged IIIa and none of them underwent surgery. Three suffered from local progression while the sites of distant relapse in the 3 remaining patients were bone, pleura and Central Nervous System.
All responses but one were obtained after the first 2 cycles. Furthermore, all unresponsive patients but one showed a disease progression at the first disease assessment.

Survival

At present, with a median follow-up of 13 mos, 2y survival of all 50 pts and of the 24 pts staged IIIa that underwent surgery is estimated as 37% (95%CI: 24-58%) and 47% (95%CI:27-80%), respectively. Results are not mature to evaluate a survival impact.

Discussion

A major randomized trial showed that survival was significantly improved with continuous hyperfractionated accelerated radiotherapy (CHART) when compared with conventional radiotherapy (RT) in patients with non-small-cell lung cancer (NSCLC) localized to the chest [11]. Taking into account that systemic chemotherapy (CT) improves survival when added to RT for patients with stage-III NSCLC [12-15], we started a prospective phase-II study with the aim to optimize a multidisciplinary approach to treat stage-IIIa/IIIb disease with a sequential combination of chemotherapy and a three-fraction-per-day accelerated radiotherapy (hfRT).
Gemcitabine (GCB) plus cisplatin (CDDP) combination (GC) was selected as induction CHT because of its confirmed antitumor activity in patients with advanced NSCLC and its proved feasibility and activity also in the induction setting [16].
Chemotherapy was scheduled s a 4-week regimen, consisting of 3 courses of CDDP 100 mg/m² d1 plus GCB 1000 mg/m² dd1,8,15, with CDDP administered after GCB. An interim analysis of toxicity was performed after the first 10 pts completed induction chemotherapy. It resulted that the treatment-related toxicity, mainly thrombocytopenia, required a frequent dose omission or reduction (90% of patients, 65% of cycles) of the GCB doses on day 15, G2 nephrotoxicity also causing chemotherapy discontinuation in 1 pt. Grade 3/4 thrombocytopenia was the main hematological toxicity, occurring in 21% of cycles.
Afterwards, the protocol was amended and the 40 pts subsequently enrolled received 119 cycles of a treatment scheduled in GCB 1250 mg/m² on days 1 and 8 and CDDP 80 mg/m² on day 1 of a 21-day treatment cycle.
Patient compliance to the three-week schedule was excellent, with a favorable toxicity profile. Dose-reduction or treatment omission were required in only 17% of patients/10% of cycles, with no treatment discontinuation. Grade 3/4 thrombocytopenia was observed in 4% of cycles, not associated with bleeding events and with recovery within a week.
The actual delivered dose intensities of both CDDP and GCB calculated to the start dose were 86.5% and 99.55% of the intended dose level with the four-week and the three-week schedule, respectively.
Our results sustained a better compliance for a three-week than a four-week schedule of GC.
The high percentage of GCB day-15 omission required with the 28-day schedule is in line with observations from studies performed both in patients with localized and with advanced NSCLC.
Many different four-week GC combinations tested in patients with advanced NSCLC showed the omission of GCB on day15 to be a limitation of the treatment, whether giving CDDP early in the cycle [4], or fractionating CDDP on days 1 and 8 [17]. Although an analysis of phase-II trials of four-week schedule showed that the administration of CDDP on day 15 required a lower percentage of GCB dose-omissions than early CDDP administration, omissions occurred in up to 29% of patients in any case, being moreover associated with a higher incidence of severe neutropenia. Also, when a four-week schedule of GC was given by EORTC as induction chemotherapy in patients with stage IIIa pN2 NSCLC, hematological toxicity frequently caused dose omission (34%) or reduction (23%) of the GCB doses on day 15, toxicity causing treatment discontinuation in 15% of patients [16]. Of interest, a three-week schedule was discussed by the authors as a possible way to reduce toxicity.
On the other hand, a phase-II randomized trial showed a three-week schedule of GC to be burdened by low toxicity, whether high (100 mg/m²) or intermediate (70 mg/m²) doses of CDDP were given. Namely, only 4% and 6% of GCB doses were reduced and omitted, respectively, when CDDP was given at the dose of 70 mg/m² [17]. A 21-day GC schedule was tested also by Cardenal in a phase-III study [5], and provided a significantly higher response-rate and delay in disease progression than an Etoposide-CDDP (EP) combination in patients with advanced NSCLC. In this study G3/4 febrile neutropenia was more pronounced in EP arm and G3/4 thrombocytopenia in the GC arm.
In our study we obtained a high response rate, with 37 partial responses (overall response: 74%), 7 stable diseases and 6 disease progressions. Although mediastinal nodes were tumor-free in only 1 out of the 24 patients who underwent surgery, the 2y survival of all the 50 pts and of the 24 pts staged IIIa who underwent surgery was estimated as 37% and 47% respectively, with a median follow-up of 13 mos.
The response rate of 74% was among the highest for induction chemotherapy, and was obtained despite all patients had a significant N2 disease at CT scan or a stage IIIb disease. Although the design of the study and the limited number of patients do not allow definitive conclusions, there was no suggestion of a different activity between the four-week and the three-week schedules, that achieved 6 out of 10 and 31 out of 40 partial responses and 1 and 5 tumor progressions, respectively. In support of this observation, the clinical response obtained in this trial is quit similar to that reported by the EORTC Lung Cancer Cooperative Group (70% of 38 patients who received the planned treatment), where only patients staged IIIa pN2 were enrolled. Similar responses have been described in the recent report of A. Depierre et al. [21]. This study was powered to investigate if there was a difference from 45% to 60% in 2-year survival rates of patients with early-stage NSCLC treated with an induction chemotherapy followed by surgery versus a control group of patients receiving primary surgery. The observed results 52% to 59% was not significant. LThe response rates reported in Depierre’s study among 179 patients treated with chemotherapy were 11% pathological complete responses and 53% partial responses. Although in the EORTC trial half of the 17 pts who underwent surgery were reported having tumor-free mediastinal nodes after induction chemotherapy [16], such a difference in pathological response should be interpreted with caution, both because of the low number of patients pathologically evaluated, and because of a possible different selection in patients that underwent surgery despite a disease stabilization after induction chemotherapy. Namely, 1 pt with no-change was randomized in the EORTC trial and 4 pts with no change were operated (2 pt: lobectomy and 2 pts: pneumonectomy) in our study.
Moreover, it should be noted that the EORTC schedule was quite similar to the schedule given to the first 10 pts treated in our study, where only one nodal pathological complete remission was obtained.
Because a GCB-related acute respiratory distress syndrome (RDS) has being recognized [18-20], acute and late pulmonary toxicity should be investigated as a possible limiting toxicity for an induction GCB-containing CHT in patients expecting to undergo sequential radiation to the lung.
In this trial the toxicity of the post-chemotherapy three-fraction-per-day accelerated radiotherapy was mild, and never required treatment discontinuation or interruption. Specifically, grade 3-G4 toxicity according to RTOG scale was registered in 3 pts (grade 3 esophagitis); no patients suffered from grade 2-4 lung toxicity, and no late clinical evidence of worsening of pulmonary function occurred. The limitation of a not-randomized trial notwithstanding, we suggest that GC combination is highly active and should be considered for induction in stage IIIa/IIIb NSCLC, and that a three-week should be preferred to a four-week schedule because of its better toxicity profile.

References

1. Luud B, Kristjanson PEG, Hansen HH. Clinical and preclinical activity of 2',2'-difluorodeoxycytidine (gemcitabine). Cancer Treat Rev 1994 ; 19 : 45-55.

2. Bergman AM, Ruiz van Haperen VWT, Veerman G, et al. Synergistic interaction between cisplatin and gemcitabine in vitro. Clin Cancer Res 1996 ; 2 : 521-30.

3. Peters GJ, Ruiz van Haperen VWT, Bergman AM, et al. Preclinical combination therapy with gemcitabine and mechanisms of resistance. Semin Oncol 1996 ; 23 : 16-24.

4. Crinò L, Conte P, De Marinis F, et al. A randomized trial of gemcitabine cisplatin versus mitomycin, ifosfamide and cisplatin (MIC) in advanced non-small cell lung cancer (NSCLC) : A multicenter phase III study. J Clin Oncol 1999 ; 17 : 3522-30.

5. Cardenal F, Paz López-Cabrerizo M, Antón A, et al. Randomized phase III study of gemcitabine-cisplatin versus etoposide-cisplatin in the treatment of locally advanced or metastatic non-small cell lung cancer. J Clin Oncol 1999 ; 17 : 12-8.

6. Crinò L, Scagliotti G, Marangolo M, et al. Cisplatin-gemcitabine combination in advanced non-small cell lung cancer : A phase II study. J Clin Oncol 1997 ; 15 : 297-303.

7. Abratt RP, Bezwoda WR, Goedhals L, et al. Weekly gemcitabine with monthly cisplatin : Effective chemotherapy for advanced non-small cell lung cancer. J Clin Oncol 1997 ; 15 : 744-9.

8. Rinaldi M, Crinò L, Scagliotti GV, Mosconi AM, De Marinis F, Gridelli C, et al. A three-week schedule of Gemcitabine-cisplatin in advanced non-small cell lung cancer with two different cisplatin dose levels : a phase II randomized trial. Ann Oncol 2000 ; 1 : 1295-300.

9. Castellano D, Lianes P, Paz-Ares L, Hidalgo M, Guerra JA, Gomez-Martin C, et al. A phase II study of a novel gemcitabine plus cisplatin regimen administered every three weeks for advanced non-small-cell lung cancer. Ann Oncol 1998 ; 9 : 457-9.

10. World Health Organization : WHO Handbook for Reporting Results of Cancer Treatment. WHO Offset Publication n^o 48. Geneva, Switzerland, World Health Organization, 1979.

11. Saunders M, Dische S, Barrett A, et al. Continuous hyperfractinated accelerated radiotherapy (CHART) versus conventional radiotherapy in non-small cell lung cancer : a randomised multicentre trial. Lancet 1997 ; 350 : 161-5.

12. Dillman R, Herndon J, Seagren S, et al. Improved survival in stage III non-small cell lung cancer. Seven year follow-up of CALGB 8433 trial. J Natl Canc Inst 1996 ; 88 : 1210-5.

13. Sause W, Scott C, Taylor S, et al. Radiation Therapy Oncology Group (RTOG) 88-08 and Eastern Co-operative Oncology Group 4588 : preliminary results of a phase III trial in regionally advanced, unresectable non small cell lung cancer. J Natl Canc Inst 1995 ; 87 : 198-205.

14. Non-Small Cell Lung Cancer Co-operative Group : Chemotherapy in non-small cell lung cancer : a meta-analysis using updated data on individual patients from 52 randomized clinical trials. Br Med J 1995 ; 331 : 899-909.

15. Pritchard RS, Anthony SP. Chemotherapy plus radiotherapy compared with radiotherapy alone in the treatment of locally advanced, unresectable, non-small-cell lung cancer. A meta-analysis. Ann Intern Med 1996 ; 125 : 723-9.

16. Van Zandwijk N, Smit EF, Kramer GW, Schramel F, Gans S, Festen J, et al. Gemcitabine and cisplatin as induction regimen for patients with biopsy-proven stage IIIA N2 non-small-cell lung cancer : a phase II study of the European Organization for Research and Treatment of Cancer Lung Cancer Cooperative Group (EORTC 08955). J Clin Oncol 2000 ; 18 : 2658-64.

17. Huisman C, Giaccone G, van Groeningen CJ, Sutedja G, Postmus PE, Smit EF. Combination of gemcitabine and cisplatin for advanced non-small cell lung cancer : a phase II study with emphasis on scheduling. Lung Cancer 2001 ; 33 : 267-75.

18. Pavlakis N, Bell DR, Millward MJ, Levi JA. Fatal pulmonary toxicity resulting from treatment with gemcitabine. Cancer 1997 ; 80 : 286-91.

19. Brose MS, Cheng JM, Haller DG. Incidence of pulmonary and non-hematologic toxicities related to gemcitabine. Proc Am Soc Clin Oncol 2000 ; 19 : 196 (abstract).

20. Tempero MA, Brand R. Fatal pulmonary toxicity resulting from treatment with gemcitabine. Cancer 1998 ; 82 : 1800-1.

21. Depierre A, Milleron B, Moro-Sibilot D, Chevret S, Quoix E, Lebeau B, et al. from the French Thoracic Cooperative Group. Preoperative chemotherapy followed by surgery compared with primary surgery in resectable stage I (Except T1N0), II, and IIIa non – small-cell lung cancer. J Clin Oncol 2002 ; 20 : 247-53.