LUNG F CANCER

Lung Cancer 10 (1993) 73-84

Non-resectable Stage IIIa-b lung carcinoma: a Phase II study on continuous infusion of

cisplatin and concurrent radiotherapy (plus adjuvant surgery)

Amedeo V. Bedini*, Luca Tavecchio, Franc0 Milani, Albert0 Gramaglia, Carlo Spreafko, Alfonso Marchianb,

Gianluigi Ravasi

Istituto Nazionale per lo Studio e la Cura dei Tumori, Via G. Venezian I, 20133 Milan, Italy

(Received 21 October 1992; revision received 21 December; accepted 1 January 1993)

Abstract

Thirty-eight patients with non-resectable non-small-cell Stake Ha-b lung cancer were treated in a Phase II study with radiotherapy (50 Gy in a 25-fraction split-course) plus con- current continuous infusion of cisplatin given at a daily dose of 6 mglm2, with the aim of investigating its radiopotentiation properties. Treatments were given on an outpatient basis by means of a central venous catheter and a portable pump. Adjuvant surgery was undertaken when feasible. Toxicity was mild to moderate. The probability of a partial or complete locoregional response at 4 weeks after treatment completion was 83% (confidence limits at 95%: 13). Eighteen patients were resected. Overall l-, 2- and 3-year progression-free survival probabilities were 42, 24 and 21%. These figures were 63, 37 and 24% in observed survival curves. Patients with squamous-cell tumors had observed survival rates of 82, 50 and 28% at 1, 2 and 3 years, compared to 42, 19 and 19% in patients with non-squamous histology. The high response and survival rates obtained at a low price according to toxicity require further investigation.

Key work Non-small-cell lung cancer; Cisplatin; Continuous infusion; Radiotherapy; Adjuvant surgery

??Corresponding author.

0169-5002/93/$06.00 0 1993 Elsevier Science Ireland Ltd. All rights reserved. SSDI 0169-5002(93)00236-3

74 A. V. Bedini et al. /Lung Cancer 10 (1993) 73-84

1. Introduction

Local failure in patients submitted to high-dose radiotherapy (RT) for inoperable lung carcinoma is exclusive in about l/3 of cases, both at clinical observation [5,14], and at autopsy findings [16]. The achievement of an effective locoregional control should therefore result in a better prognosis [12]. Half of the patients submitted to 50-60 Gy for non-operable lung carcinomas achieve an objective response. The median survival is reached in 8- 12 months, and the 2-year survival rate is about 15% [5,6,13,17,19]. However, no survival advantage is appreciated following high-dose RT administration [6], whose employment in clinical practice as routine has been strongly questioned [ 111. Four Phase III studies investigated the radiopotentiation properties of cis-dichloro-diammine-platinum (CDDP) given concurrently with RT [ 1,17,20,23]. Various bolus doses with different timing of delivery, combined with different schedules of RT were evaluated. Since the modality of CDDP administra- tion seems crucial in achieving optimal radio-enhancement, we designed a Phase II study on continuous infusion of CDDP concurrently with high-dose RT. Adjuvant surgery was undertaken when feasible.

2. Materials and methods

2. I. Eligibility Thirty-eight patients (32 males, mean age 58 years, ranging from 39 to 74) with

clinically non-resectable Stage IIIa-b [9] non-small-cell lung carcinoma, entered the study at the National Cancer Institute of Milan (May 1988-October 1989). T4 or N3 assessment, or the presence of at least one mediastinal node of diameter > 3 cm, or involvement of three or more nodal stations, were criteria for non-resectability. Nodes were considered positive if their largest diameter was 15 mm or more. Nodal mapping followed the American Thoracic Society (ATS) rules [22]. Stage assessment was based upon physical examination, chest X-rays, bronchoscopy, CT or MR stud- ies of brain, chest, and upper abdomen. Bone scan was performed only when skeletal pain and/or tenderness, alkaline phosphatase exceeding 108 IU/dl, hypercalcemia, were present. Entry criteria included a pathological diagnosis, a O-2 Eastern Cooperative Oncology Group (ECOG) performance status (PS) [lo], normal blood cell counts and renal function, informed consent. Patients with pleural effusion, su- perior vena cava syndrome, scalene/supraclavicular N3 spread, age > 75 years, pre- vious chemo- or radiotherapy, RT given by other institutions were excluded. Standard criteria ruled the assessment of concurrent functional inoperability [7]. Registration in the study followed RT simulation.

2.2 Treatment The design of the study is shown in Fig. 1. Daily doses of 2 Gy (adjustments in

the range of 1.8-2.5 Gy were admitted) were given 5 days a week, in the first 3 and last 2 weeks of a 7-week split-course: the total dose was 50 Gy. Two linear accelerators (6 and 15 MeV) and a betathron unit (45 MeV) were used. Tumor and involved nodes were encompassed by 2-cm safe margins. The sites routinely irradiated were the mediastinum, with the lower margin 5 cm below the carina, the

A. V. Bedini et al. /Lung Cancer 10 (1993) 73-84 15

CDDP INFUSION 6 MGIMP DAILY

AAAAA FIT-2 GY DAILY

NON-RESECTABLE STAGE WA-B

LUNG CARCINOMA

RESECTABILITY RESECTABILITY

7 OR

NON-N3-NON-Ml DJUVANT SURGERY STAGE

Fig. 1. Design of the study.

contralateral hilum, and both supraclavicular fossae. Almost the whole treatment was delivered by means of two opposed portals, blocking the spinal cord at 38 Gy (i.e. following the 19th session) because of an expected increased risk of damage due to CDDP. Both the supraclavicular fossae were shielded at 40-45 Gy, but the prescribed dose was reached. within the 20th session, due to the lower thickness of these sites. A third, occasionally a fourth portal, encompassing the whole radiologically assessable disease, was thereafter planned for treatment delivering by means of a betathron unit. A tunnelled Groshong central catheter (Catheter Tech- nology Co., Salt Lake City, UT) was inserted via a subclavian vein. Continuous infu- sion of CDDP (Platinex solution, Bristol-Myers, Madrid, Spain) at a daily dose of 6 mg/m2 was started before each first weekly RT fraction and suspended about 120 h later. Drug delivery was guaranteed by a portable pump (Cadd-1, Pharmacia Deltec Inc., St. Paul, MN). A physical examination twice a week during treatment, and blood cell counts and renal function at weeks 2, 3, 6, 7, and 11 were performed for toxicity assessment. CDDP was lowered 20-30% for grade 2 leukopenia, and in- terrupted for grade 1 nephrotoxicity or any other toxicity of grade 3 according to World Health Organization (WHO) rules [8]. Therapy was administered on a full outpatient basis and no supportive care was given as routine. However, patients were recommended to have a daily oral fluid intake of at least 1.5-2 1 under treatment.

CT or MR scans of brain, chest, and upper abdomen were taken at weeks 5, and 11, for response evaluation. It was initially stipulated that if the aforementioned non-

76 A. V. Bedini et al. /Lung Cancer 10 (1993) 73-84

resectability and non-operability criteria were not met at these restagings, a resection would be planned. However, after early observation of unexpected relevant responses, or gross underestimation of response, surgery was proposed for all pa- tients with clinical non-N3-non-Ml disease, if functionally operable, following restaging at week 11 even though the resectability criteria were not fulfilled.

2.3. End points of the study The radiographic disappearance of irradiated disease was defined as complete

locoregional response (CR); a >50% reduction of the product of the two greatest perpendicular diameters of the tumor, or a > 50% decrease in the sum of the prod- ucts of the perpendicular diameters of all the lesions, if multiple lesions were present (tumor, nodes), were defined as partial locoregional response (PR). CR plus PR were considered major responses. The extent of locoregional disease was stable (SD) if the decrease was less than 50% or absent. Increased product or sum of products of any degree, or the occurrence of new lesions within the irradiated area were considered as locoregional or intrafield progression. Progression was coded as extratield or dis- tant if new lesions occurred outside the irradiated field. The 5- and 1 l-week pro- babilities of a locoregional major response were calculated according to the actuarial method previously described [3]. No patient was excluded. CT/MR/other roen- tgenological documentation had a second-party review for tumor measurements and response assessment by experienced radiologists.

Observed and progression-free survival curves were computed following the life-table method (the 1st day of treatment being time zero), having death from all causes, or progression/death not due to cancer as end-points, respectively. The time of analysis was 1 October 1992. Complete information on outcome was available for all patients, whose clinical records were reviewed by the study chairman (GR) and the data manager (LT).

3. Results

3.1. Patient characteristics There were 22 squamous-cell cases, 13 adenocarcinomas, and three tumors of

other types. Eleven patients (or 29%) had a Stage IIIa tumor at registration, the re- maining having Stage IIIb disease due to a T4 assessment (17 patients, nine of whom with N2 disease also), or N3 spread (seven patients) or both these conditions (three patients). Two patients had obtained down-staging from Stage IV after resection of a solitary brain metastasis. Thirty patients were graded as O-l, seven as 2, and one as 3, according to PS. Thirty-seven were symptomatic, live with a weight loss > 5%, seven were medically non-operable, and five had suffered of a previous malignancy. Thirty-two (84%) primary tumors had one diameter >30 mm, and 17(45%) >60 mm, i.e. the cut-off values believed to be of negative prognostic impact [15]. Thirty patients were diagnosed with positive nodes, and 10 of them underwent surgical stag- ing. The total number of positive nodal stations was 87 (34 surgically assessed). The median number of involved ATS stations per patient was three (range: l-6). The median diameter of positive nodes was 28 mm (range: 16-70 mm).

A. V. Bedini et al. /Lung Cancer 10 (1993) 73-84 II

3.2. Protocol violations There was one Stage IV tumor, underestimated at registration. One patient had

a T3 NO tumor (registration stage: T4 NO); one further patient, although accrued after an exploratory thoracotomy for technically non-resectable N2 disease, had only one mediastinal station involved, with nodes less than 3 cm of diameter at CT scan. Moreover, one patient had an ECOG performance status of 3, and four receiv- ed an initial daily CDDP dose of 5 mg/m*, the reasons for the reduction being age over 70 years (two patients), a creatinine clearance value of 59 ml/min (one patient), and previous occurrence of altered renal function (one patient).

3.3. Devices and toxicity The reliability of the devices used was described in detail in two previous papers

[2,3]. Less than 1% of the predicted time of infusion was lost due to catheter- or pump-related complications. A summary on given therapy is shown in Table 1. All the patients were evaluated for toxicity. CDDP-related toxicity, occurred within 11 weeks, is shown in Table 2. A mild esophagitis was observed in almost all patients, according to early toxicity due to RT. No case of radiation pneumonitis was seen. We did not observe any case of late toxicities. However, because of resection in half of the patients, and early progression or death in a further l/3, any meaningful evalu- ation is prevented as regards this issue.

3.4. Response The probabilities of a major locoregional response are outlined in Table 3.

Response rates of squamous cell cases were slightly higher, but four of the 16 patients with non-squamous histology did not undergo treatment completion for early distant progression. No locoregional progression was seen. There were three cases of locoregional Stage 0, seven of Stage I, one of Stage II, 13 of Stage IIIa, and 14 of Stage IIIb at the last clinical restaging performed according to the protocol rules.

Table 1 Summary of treatment

Week Number of treated patients

RT CDDP dose

100% 70-80% 0%

Number of interrupted treatments for

Pro. Surgery

1 38 34 4 0 0 0 2 38 34 4 0 0 0 3 31 32 4 1 I 0 6 32 25 I 0 3 2 I 32 23 6 3 0 0

78 A. V. Bedini et al. /Lung Cancer 10 (1993) 73-84

Table 2 Frequency of CDDP-related toxicity (see text for RT toxicity)

Toxicity WHO toxicity grade Total

0 1 2 3 4

Leukopenia 23 13 2 - - 38 Nephrotoxicity 32 6 - - - 38 Nausea/vomiting 14 II 12 1 - 38

3.5. Surgery Twenty-one patients with non-N3-non-Ml spread underwent surgery (14 of them

having achieved clinical resectability also), the remaining 17 being excluded for distant metastases, persistent N3 disease, functional inoperability, or multiple contraindications. However, in overall 19 patients clinical resectability was obtained. The resection rates, according to the clinical assessment of resectability, are shown in Table 4. Eleven lobectomies and six pneumonectomies were performed with stapl- ing techniques. One right upper sleeve lobectomy was also performed. Surgical pro- cedures required very high skill mainly due to post-treatment fibrosis. Thirteen of the 22 patients with squamous cell carcinoma and five of the 16 of those with other histological types were resected. Two fatal postoperative complications occurred, the former due to adult respiratory distress syndrome after a lobectomy plus chest wall resection in the T3 NO patient, the latter likely due to broncho-vasal fistula, in the patient submitted to sleeve lobectomy (an acute massive hemophtysis happened 35 days after surgery, 27 days after discharge). We registered, as non-fatal complica- tions, seven cases of pneumonitis. No further complication was seen. The mean time of hospitalization after surgery was 10.9 days (range: 6-18).

3.6. Pathological assessment Pathological examination of the primary cancer showed an absence of viable

tumor in one case, microscopic foci of disease in 10, various degrees of necrosis in seven. Thirteen patients of the 18 resected had mediastinal nodal spread before treat-

Table 3 Percent probability of a clinical major locoregional response at the restagings (95% confidence limits), in relation to T-N descriptors

Stage descriptor

Time of restaging

Fifth week Eleventh week

T 64 (15) 88 (11) N 58 (18) 79 (16) TN 61 (15) 83 (13)

A. V. Bedini et al. /Lung Cancer 10 (1993) 73-84 19

Table 4 Resection rates in relation to the clinical assessment of resectability

Clinical resectability criteria

Type of operation Total

Radical Non-radical Explorative resection resection thoracotomy

Present 12 1 I 14 Absent 3 2 2 I Total 15 3 3 21

ment with a total of 30 ATS stations (10 with surgical staging also) involved. Pathological results are shown in Table 5. The correlation between preoperative clinical versus surgical/pathological stage is displayed in Table 6.

3.7. Outcome Two of the 20 unresected patients were in clinical locoregional CR, the remaining

18 still having evaluable disease, at the start of follow-up. In two of these 18 a locoregional CR was assessed within 4 months from treatment completion. No post- operative therapy was given. The pattern of failure is shown in Table 7. Twelve ini- tial distant and two locoregional progressions occurred within 6 months (P < 0.01). Three patients with squamous cell carcinoma of the 11 at risk after resection (the two postoperative deaths were excluded) had initial progression at locoregional sites, and so had four of the nine who did not undergo resection. Locoregional failure oc- curred in one resected and in one non-resected patient among the non-squamous cell cases. Progression-free survival curves are displayed in Fig. 2, and those of observed survival in Fig. 3. The l-, 2-, 3-year survival rates of the 15 radically resected patients were 80,60 and 47%. These figures were 75,42 and 17% for the 12 patients who were assessed as non-N3-nonM1 at the 11-week restaging, but did not undergo surgery for functional non operability (six patients), or were explored (three patients), or non radically resected (three patients).

Table 5 Results of pathological examination of nodal stations assessed as involved before treatment

Modality of pretreatment staging

Number of nodal stations examinated Total (pathological findings)

Complete Microscopic Regressive response disease features

Clinical 1 3 10 20 Surgical 5 2 3 IO Total 12 5 13 30

80 A. V. Bedini et al. /Lung Cancer 10 (1993) 73-84

Table 6 Comparison of preoperative clinical staging and surgical/pathological staging

Surgical/ Clinical stage Total pathological

0 I II III

0 - - - 1” 1 I - 3c - 6” 9 II - - lC 2” 3 III lb 2b - 5c 8 Total 1 5 1 14 21

Vlinical overestimation of S/p stage: 9/21 (43%). bUnderestimation: 3/21 (14%). CConcordance: 9/21 (43%).

4. Discussion

The radiopotentiation properties of CDDP were investigated in four randomized studies on non-operable non-small cell carcinoma. Soresi et al. [20] evaluated 50.4 Gy given in 28 fractions in 5.5 weeks, or the same plus CDDP (15 mg/m* once a week during RT administration), obtaining a higher rate of major responses, a long- er median survival, and a lower frequency of local progression in the combined- therapy arm. Schaake-Koning and the EORTC group [17] randomized 331 patients

Table 7 Initial modality of failure

Modality Histology

squamous carcinoma (N = 22)

Non-squamous carcinoma (N= 16)

Total (N = 38)

Extrafield progression Brain Bone Nodes Lung Multiple Sites

Intratield progression ‘T ‘N

Combined modalities Non cancer-related deaths Total

7 2 1 2 2

- 7 4 3 1 28

17

10 17 4 6 4 5 1 3

- 2 1 1 2 9 2 6

- 3 - 1

lb 3 13 30

BPostoperative deaths. bMyocardial infarction.

A. V. Bedini et al. /Lung Cancer IO (1993) 73-84 81

%

S U R V

I

V I N G

%

S U R V I V I

N

G

100

90

80

70

60

50

40

30

20

10

100

90

80

70

60

50

40

30

20

10

-4 ‘I

_ I _ OVERALL - I

- L __ SQUAMOUS

t- _ NON-%.

I I I 1

1 2 3

TIME TO PROGRESSION (YEARS)

Fig. 2. Progression-free survival.

OVERALL

_ _SQUAMOUS ‘i,,

I -NON-SQ.

Median

(Mos)

7

16

1 2 3

TIME TO DEATH (YEARS)

Fig. 3. Observed survival.

4.5

Median % (Mos) 1Yf

14 63

23 62

8 38

% 1Yr

42

55

25

Surviving

2Yr 3Yr

24 21

27 23

19 19

Survlving

2Yr 3Yr

37 24

50 28

19 19

82 A. V. Bedini et al. /Lung Cancer IO (1993) 73-84

to receive 55 Gy in a 6-week split-course of 20 fractions, or the same plus CDDP given in a weekly bolus dose of 30 mg/m2, or given in a daily bolus dose of 6 mg/m2 administered just before each radiation fraction. Locoregional control and survival were significantly improved in patients of the 3rd arm, and an overall positive trend in favour of combined therapies was seen. Ansari et al. [ 1] did not observe any thera- peutic advantage from three CDDP administrations of 70 mg/m2 every 21 days dur- ing a 60-Gy RT course with standard fractionation, nor did Trovo’ et al. [23] from 15 CDDP deliveries of 6 mg/m2 given 30-60 min before each 3-Gy fraction of a 45 Gy continuous course.

Some consideration should be given to the optimization of CDDP and RT schedules. CDDP is currently delivered in bolus or rapid infusion, having initial t1/2s values in the decay curves of about 30 min. More than 90% of total plasma CDDP is protein bound within 2 h after administration, and the free CDDP may be almost completely eliminated from the plasma within 3 h. However, exposure to free CDDP species as measured by the area under the curve (AUC) is increased by prolonged infusions [25]. CDDP seems to potentiate radiation cell killing by two distinct but not exclusive modalities [4]. The former is believed to be activated by a fast free radi- cal or radiation chemical process, but the plasma concentration of unbound CDDP needed is reached only soon after administration of bolus doses of 20-100 mg/m2. The latter involves slow biochemical or radiation biochemical mechanisms, in- hibiting repair of radiation damage, requiring lower concentrations. The interaction of CDDP with radiation, as it may occur in clinical practice, should hence involve this radiation biochemical-based mechanism. On the optimal time interval and the best sequence of administration of CDDP and radiation only preclinical data are available, showing that the growth delay of squamous cell carcinoma spheroids is enhanced both by the dose of RT and its multiple fractionation and by the concur- rent administration of CDDP versus pre- or post-treatment administration [ 181. Pro- longed infusion of CDDP given concurrently with RT seems therefore the most attractive schedule to reach radio-enhancement, with an expected decrease of toxic effects [25]. Increased RT fractionation seems preferable as well.

This study was designed with the aim of evaluating the effect of CDDP delivering given according to the best-at least from our point of view-theorical schedule, with concurrent RT. However, a split-course RT was planned to lower the CDDP dose- intensity, and prevent an otherwise expected severe toxicity (241.

The use of modern devices for prolonged venous access maintenance and of portable infusion pumps allow this approach be highly realiable, and safe. We observed locoregional response rates higher than those usually reported in the litera- ture, despite the prevalence of unfavorable prognostic factors (e.g. stage or tumor diameter), with a very low price as regards to toxicity. In our opinion, these results cannot be explained by a selection bias. Surgery was feasible without appreciable increase of risk, but no conclusion can be drawn on its role on prognosis, since this study was not especially focused on resection. The post-treatment clinical assessment is only roughly reliable, since often underestimates response and overestimates non- resectability, but is highly predictive for resectability. Patients at enhanced risk of locoregional progression, i.e. those with squamous histology [2 11, showed a twofold longer median survival than usually reported in literature. This schedule of treat-

A. V. Bedini et al. /Lung Cancer 10 (1993) 73-83 83

ment deserves further investigation. New pilot and Phase II studies are currently in progress with the aim both of optimizing CDDP doses and RT courses, and of investigating adjuvant surgical approach.

5. Acknowledgments

This work was partially supported by a grant of the Italian League against Cancer- Milan Division. We wish to thank Dr M. Valente for his valuable suggestions.

6. References

1

2

3

4

5

6

7

8

9 10

11

12

13

14

15

16

17

18

Ansari R, Tokars R, Fisher W et al. A phase III study of thoracic irradiation with or without concomitant cisplatin in locoregional unresectable non-small cell lung cancer (NSCLC): a Hoosier Oncology Group (H.O.G.) protocol [abstract]. ASCO Proc 1991; IO: 823. Bedini AV, Tavecchio L, Bonalumi MG et al. Reliability of prolonged infusion in cancer chemother- apy with the Groshong central venous catheter. Reg Cancer Treat 1990; 3: 232-234. Bedini AV, Tavecchio L, Milani F et al. Prolonged venous infusion of cisplatin and concurrent radiation therapy for lung carcinoma. A feasibility study. Cancer 1991; 67: 357-362. Coughlin CT, Richmond RC. Biologic and clinical developments of cisplatin conbined with radiation: concepts, utility, projections for new trials, and the emergence of carboplatin. Sem Oncol 1989; 16 Suppl 6: 31-43. Curran WJ, Stafford PM. Lack of apparent difference in outcome between clinically staged IIIa and IIIb non-small-cell lung cancer treated with radiation therapy. J Clin Oncol 1990; 8: 409-415. Johnson DH, Einhorn LH, Bartolucci A et al. Thoracic radiotherapy does not prolong survival in patients with locally advanced, unresectable non-small cell lung cancer. Ann Intern Med 1990; 113: 33-38. Khalil Ah M, Ewer MS. Preoperative cardiopulmonary evaluation of patients undergoing surgery for lung cancer. Cancer Bull 1980; 32: 100-104. Miller AB, Hoogstraten B, Staquet M et al. Reporting results of cancer treatment. Cancer 1981; 47: 207-214. Mountain CF. A new international staging system for lung cancer. Chest 1986; 89 Suppl: 225-232. Oken MM, Creech RH, Tormey DC et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol 1982; 5: 649-655. Payne DG. Non-small-cell lung cancer: should unresectable stage III patients routinely receive high- dose radiation therapy? J Clin Oncol 1988; 6: 552-558. Perez CA, Bauer M, Edelstein S et al. Impact of tumor control on survival in carcinoma of the lung treated with irradiation. Int J Radiat Oncol Biol Phys 1986; 12: 539-547. Perez CA, Bauer M, Emami BN et al. Thoracic irradiation with or without levamisole (NCS 177023) in unresectable non-small cell carcinoma of the lung: a phase HI randomized trial of the RTOG. Int J Radiat Oncol Biol Phys 1988; 15: 1337-1346. Perez CA, Pajak TF, Rubin P et al. Long-term observations of the patterns of failure in patients with unresectable non-oat cell carcinoma of the lung treated with definitive radiotherapy, Cancer 1987; 59: 1874-1881. Sandler HM, Curran WJ, Turrisi AT. The influence of tumor size and pre-treatment staging on out- come following radiation therapy alone for stage I non-small cell lung cancer. Int J Radiat Oncol Biol Phys 1990; 19: 9-13. Saunders Ml, Bennet MH, Dische S et al. Primary tumor control after radiotherapy for carcinoma of the bronchus. Int J Radiat Oncol Biol Phys 1984; IO: 499-501. Schaake-Koning C, van den Bogaert W, Dalesio 0 et al. Effects of concomitant Cisplatin and radiotherapy on inoperable non-small-cell lung cancer. New Engl J Med 1992; 236: 524-530. Schwachofer JHM, Crooijmans RPMA, Hoogenhout J et al. Effectiveness in inhibition of recovery of cell survival by cisplatin and carboplatin: influence of treatment sequence. Int J Radiat Oncol Biol Phys 1991; 20: 1235-1241.

84 A. V. Bedini et al. /Lung Cancer 10 (1993) 73-84

19 Simpson JR, Bauer M, Perez CA et al. Radiation therapy alone or combined with misonidazole in the treatment of locally advanced non-oat cell lung cancer: report of an RTOG prospective rando- mized trial. Int J Radiat Oncol Biol Phys, 1989; 16: 1483-1491.

20 Soresi E, Clerici M, Grilli R et al. A randomized clinical trial comparing radiation therapy v radia- tion therapy plus cis-dichlorodiammine platinum (II) in the treatment of locally advanced non-small cell lung cancer. Sem Oncol 1988; 15 Suppl 7: 20-25.

21 Stanley K, Cox JD, Petrovich Z et al. Patterns of failure in patients with inoperable carcinoma of the lung. Cancer 1981; 47: 2725-2729.

22 Tisi GM, Friedman PJ, Peters RM et al. Clinical staging of primary lung cancer. Am Rev Respir Dis 1983; 127: 659-669.

23 Trovb MG, Minatel E, Franchin G et al. Radiotherapy versus radiotherapy enhanced by cisplatin in stage III non-small cell lung cancer. Int J Radiat Oncol Biol Phys, 1992; 24: 11-15.

24 Van Harskamp G, Boven E, Vermorken JB et al. Phase II trial of combined radiotherapy and daily low-dose cisplatin for inoperable locally advanced non-small cell lung cancer (NSCLC). Int J Radiat Oncol Biol Phys 1987; 13: 1735-1738.

25 Vokes EE, Ackland SP, Vogelzang NJ. Cisplatin, Carboplatin and Gallium Nitrate. In: Lockich JJ, editor. Cancer chemotherapy by infusion, 2nd edn. Chicago, Precept Press 1990: 176-196.