Annals of Oncology 10: 1125-1128, 1999.© 1999 Kluwer Academic Publishers. Primed in the Netherlands.

Short report

Oxaliplatin and paclitaxel combination in patients with platinum-pretreatedovarian carcinoma: An investigator-originated compassionate-use experience

S. Faivre,1 S. Kalla,2 E. Cvitkovic,3 O. Bourdon,1 D. Hauteville,4 L. M. Dourte,4

M. A. Bensmaine,2 M. Itzhaki,2 M. Marty1 & J. M. Extra1

^Service d'Oncologie Medicate, Hopual Saint-Louis, Paris; 2CAC, Kremlin Bicetre, iHopital Paul Brousse, Villejuif,Hopital Begin, Saint-Mande, France

Summary

Purpose: Compassionate-use oxaliplatin-paclitaxel was as-sessed for toxicity and efficacy according to clinical platinumresistance status in cisplatin-carboplatin-pretreated advancedovarian cancer patients.

Patients and methods: Thirty-seven patients, retrospectivelygrouped into four oxaliplatin-paclitaxel dose levels (mg/m2):(DL1: 100/135; DL2: 130-135/135; DL3: 100/160-175; DL4:130-135/160-175), received oxaliplatin and paclitaxel everythree to four weeks.

Results: Thirty-one of thirty-seven treated patients wereevaluable for activity, with 1 complete and 14 partial responses,(objective response rate: 48%, 95% CI: 31-66). Of 18 platinum-resistant patients 6 responded, and of 13 platinum-sensitivepatients, 9 responded. One patient (3%) had two febrile neutro-

penia episodes, and eight (22%) and eleven patients (30%) hadgrades 3 and 4 neutropenia, respectively. Six patients (16%)experienced grade 3 peripheral neuropathy. The median re-sponse duration was 10.8 months, with a 23-month (range8-54) median follow-up. Median progression-free and overallsurvivals were 9 months (95% CI: 7-12), and 25.2 months(95% CI: 12-39), respectively.

Conclusions' The antitumour activity of oxaliplatin-pacli-taxel in platinum-resistant ovarian cancer patients accordswith experimental data on the agents' lack of cross-resistance.Time-related progression parameters confirm it as a promisingsalvage treatment option.

Key words: DACH-platinum, neurotoxicity, salvage chemo-therapy, taxanes

Introduction

Since the incorporation of platinum compounds intoovarian cancer (OC) treatment, disease control and sur-vival in advanced OC patients have improved. Single-agent paclitaxel is an active salvage compound, withobjective response rates (ORRs) of 14% to 33%, and itscombination with cis-/carboplatin in first-line treatmenthas improved time-related disease-progression parame-ters [1].

Oxaliplatin (Eloxatin®), is a new platinum compoundwith a mechanism of action similar to that of classicplatinum compounds, but with a different cytotoxic pro-file. It appears not to be cross-resistant with cis-/carbo-platin in human ovarian cell lines [2, 3]. Oxaliplatin hasa characteristic toxicity profile with acute cold-triggereddysesthesias and cumulative neurosensorial toxicity asits principal side effects, but without haematological,renal or auditive toxicity at recommended doses [4].Single-agent oxaliplatin is active in cisplatin-pretreatedor resistant OC, and, combined with cisplatin, displaysclinical synergy in heavily pretreated OC patients [5, 6].

Oxaliplatin and paclitaxel single-agent activity [5, 7]in cisplatin-resistant OC, and their non-overlapping

toxicity profiles make them suitable for combination [8],as was done in a compassionate-use programme withpretreated OC patients. Since both induce peripheralneurotoxicity, this side effect was carefully assessed incis-/carboplatin-pretreated patients.

Patients and methods

Patient selection

Between June 1994 and July 1998, 37 pretreated OC patients wereenrolled in an oxaliplatin individual patient allowance extended-accessprogram. Oxaliplatin was supplied by Sanofi Winthrop upon physicianrequests. Written information was given to each patient, and treatmentwas conducted under the prescribing oncologist's responsibility.

Patients were to have confirmed advanced OC, progressed/relapsedafter ^ I prior cis-/carboplatin-containing chemotherapy; PS = 2(WHO), peripheral neurotoxicity = 1 (NCI-CTC), adequate bonemarrow (neutrophil count 51500/ml. haemoglobin ^9.5 mg/dl, pla-telet count > 100,000 /ml), liver (bilirubin ^ 2x upper limit of normal(ULN), AST and ALT =£3x ULN, or ^ 5 ULN in case of livermetastasis), and renal (creatinine <2x ULN or creatinine clearance>60 ml/min) functions. As this was a compassionate-use program,disease evaluability criteria and eligibility guidelines were not strictlyadhered to.

Downloaded from https://academic.oup.com/annonc/article-abstract/10/9/1125/173398by gueston 08 February 2018

1126

Treatment

Oxaliplatin and paclitaxel were given sequentially on the same dayevery 3-4 weeks. Oxaliplatin (100-135 mg/m2). and paclitaxel (135-175mg/m2) dose levels were determined according to number and inten-sity of prior chemotherapy regimens and residual toxicities. Oxalipla-tin was administered i.v. over two or six hours, diluted in 500 ml of 5%dextrose, followed by paclitaxel as a three-hour infusion in 500 ml of5% glucose. Paclitaxel premedication was 130 mg of prednisolone peros. 12 and 6 hours before treatment administration. I.v. dexchlor-pheniramine (5 mg) and ranitidine (150 mg) were administered 30minutes before paclitaxel. Antiemetics were administered accordingto center practice. Patients were retrospectively grouped into oxali-platin-paclitaxel dose levels (DLs) (mg/m2). DL1: 100/135. DL2:130-135/135, DL3: 100/160-175, DL4: 130-135/160-175.

Toxicity and response evaluation

Toxicities were graded using the NCI-CTC scale Complete blood cellcounts were performed weekly. Serum creatinine, blood electrolytelevels and CA 125 serum levels were obtained before each treatmentcycle.

Antitumour activity (WHO-UICC) was assessed by clinical exami-nation and imaging techniques An Expert Review Committee (ERC)assessed patients reported by the treatment oncologist as having aresponse, disease stabilisation 3= 12 weeks' duration, or as non-evalu-able (treatment following optimal debulking surgery).

Progression-free survival (PFS) and overall survival (OS) werecalculated by the Kaplan-Meier method from the beginning of treat-ment to disease progression/death, respectively Survival follow-upcontinued until March 1999.

Results

Patient characteristics, treatment exposure and toxicity

Table 1 presents demographic, pretreatment, disease his-tory and extent characteristics. Nine patients had for-merly received carboplatin in high-dose chemotherapyregimens. According to Markman's criteria [9], 20 pa-tients were refractory to platinum compounds, 16 weresensitive, and 1 unclassified. All treated patients wereevaluable for toxicity, while 31 of them had measurable/evaluable targets for antitumour activity assessment.

A total of 209 treatment cycles were given (median:6 cycles/patient; range: 2-9). Overall median dose in-tensities: oxaliplatin = 48 mg/m2/week; paclitaxel = 35.3mg/m2/week. Median cumulative doses: oxaliplatin =650 mg/m2/patient (range 200-1170); paclitaxel = 960mg/m2/patient (range 300-1575).

Table 2 presents toxicity data. Haematological tox-icity was moderate, with Grade 3-4 neutropenia in 16%and 13% of cycles, respectively. One patient had twoepisodes of neutropenia-related fever. Neutropenia epi-sodes were brief. In the nine patients pretreated withhigh-dose chemotherapy (57 cycles), haematologicaltoxicity was moderate, with grade 3-4 neutropenia inthree patients.

Peripheral neurotoxicity was the most common sideeffect, seen in 94% of patients. The maximum cumula-tive neurosensory toxicities were grade 2 (12 patients)and 3 (6 patients, NCI-CTC), with four patients experi-

Table 1 Patient characteristics (n = 37).

Characteristics

Performance status (WHO)0-123Missing data

Histological typeSerousOtherMissing data

Median number of involved sites (range)0a

125=3

Involved sitesPeritoneumPelvisLymph nodesLiverPleuraOther

Median number of prior regimens123

Prior regimens containingCisplatin

Mean cumulative dose ± SDCarboplatin

Mean cumulative dose ± SDBoth cisplatin/carboplatin

Platinum-free interval (months)MedianRange< 6 - < 12 months5= 12 monthsUnclassified11

Number ofpatients

24319

24491 (0-5)4

15126

17119858

21124

33528 + 190 mg

161633 +735 mg

12

50-45

2692

%

6583

24

651124

11413216

463024221422

1573211

89

43

32

70245

a Four patients (11%) were treated after maximal debulking surgeryleaving microscopic residual disease, but no macroscopic measurableor evaluable disease.

encing grade 3 toxicity after a ^ 520 mg/m2 cumulativeoxaliplatin dose, and two patients with cumulative dosesbetween 780 and 1170 mg/m2.

Of the 28 patients (77%) with neurotoxicity symp-toms at treatment discontinuation, and available for ^ 6months follow-up, 10 (36%) recovered completely by thefirst follow-up (3-9 months).

Treatment delays and dose modifications

Eleven cycles were delayed because of haematologicaltoxicity, most of them (eight of eleven) for < 2 weeks.One patient discontinuing treatment after two cyclesbecause of grade 3 neurotoxicity (severe myalgia andnail dystrophy) recovered completely.

Eight of one hundred seventy-two (5%) cycles hadminor (15%—20%) dose reductions, mostly without spe-cific reasons.

Downloaded from https://academic.oup.com/annonc/article-abstract/10/9/1125/173398by gueston 08 February 2018

1127

Table 2. Severe (grade 3-4 NCI-CTC) toxicity. by patient and cycle.

Toxic- Number ofity per patients/dose cycles evalu-level able

Number of grade 3events (%)

Patient Cycle Patient Cycle

Number of grade 4events (%)

Patient Cycle

AnaemiaAll 37DL1 8DL2DL3DL4

NeutropeniaAll 37DL1DL2DL3DL4

96

14

896

14

19846513170

19546513167

ThrombocytopeniaAll 37 199

46513171

96

14

DL1DL2DL3DL4

VomitingAll 37DL1 8DL2 9DL3 6DL4 14

DiarrhoeaAll 37DL1DL2DL3DL4

896

14

20246533370

20547533372

1(3)

1(7)

8(22)1(12.5)1 (11)

6(43)

3(8)1(12 5)2(22)

5(13.5)2(25)

1(11)

1(17)

1(7)

1(3)

1(7)

1 (0.5)

1(1.5)

32(16)4(9)

14(27.5)3(10)

11(16)

5(2.5)3 (6.5)2(4)

5(2 5)2(4)1(2)1(3)1(1.5)

0.5

1(1.5)

11 (30)1(12.5)5(56)1(17)4(29)

25(13)1(2)

12(23.5)2 (6.5)

10(15)

1 (3) 1 (0.5)

1(11) 1(2)

For data given by dose level, values are indicated in reference topatients/cycles by dose level.

Treatment activity

Table 3 summarizes response data. According to theERC, 15 of the 37 patients had objective responses,yielding an ORR of 41% (95% CI: 25-56). In the 31evaluable patients, the ORR was 48% (95% CI: 31-66).Causes of non-evaluability were: treatment discontinua-tion after two cycles without evaluation (one patient),optimal debulking surgery (one patients), baseline icon-ography > 8 weeks before treatment (one patient),undergoing treatment at cut-off date (one patient). Forthe 37 patients treated, median response duration was10.8 months. Six of fifteen responders had received onlyone prior chemotherapy line, whereas seven and twopatients had received two and three lines, respectively.Nine of fifteen responders (60%) had ^ 2 sites involved.Objective responses were observed in 6 of 9 nodalinvolvement instances, 4 of 10 assessable pelvic sites,and 3 of 7 liver-involvement cases.

Six of eighteen platinum-resistant evaluable patients(33%) and 9 of 13 platinum-sensitive evaluable patients(69%) had objective responses.

Table 3. Response to oxaliplatin-paclitaxel combination by dose level.

Response for evaluable DL1 DL2 DL3 DL4 Totalpatients (n = 7) (;i = 9) (« = 3) (n = 12) (n=31)

Complete response -Partial response 2 (29%) 5 (56%) -Disease stabilisation 2(29%) 2(22%) 2(67%)Disease progression 3 (43%) 2 (22%) 1 (33%)

Objective response rate 2(29%) 5(56%) -95% CI (%) 0-62 23-88

1 (8%) I (3%)7(58%) 14(45%)3 (25) 9 (29%)1 (8%) 7 (23%)

8(67%) 15(48%)35_90 31-66

The median PFS in the overall population was 9.4months (95% CI: 7-12 months). With a 23-month (8-54)median follow-up, there were 19 deaths, with median OSat 25 months (95% CI: 12-39).

Discussion

Resistance to cisplatin and/or carboplatin (primary andacquired) is a major limitation in OC treatment. Weinvestigated whether full recommended doses of pacli-taxel could be safely combined with oxaliplatin in plati-num-pretreated recurrent OC patients. Haematologicaltoxicity was mild to moderate, easily manageable, anddid not require growth-factor support. At the full recom-mended doses of both agents (DL4), 43% of patientsexperienced grade 3 and 29% grade 4 neutropenia.Neurosensory neuropathy was frequent, closely corre-lated with cumulative oxaliplatin doses, but reversible.

Objective responses were seen in half the evaluablepatients in all tumour sites. The 10-month responseduration, with a PFS of similar length and a long mediansurvival are encouraging, given the pretreatment char-acteristics.

In two other studies, single-agent high-dose carbo-platin was given to patients relapsing after a platinumregimen [10, 11]. Severe haematological toxicity wasprevalent, with significant morbidity and a need forgrowth factors. The reported response rates were 38%[10], and 57% [11], but response was defined differently.

There is little data on the paclitaxel/cisplatin orcarboplatin combination in cisplatin-carboplatin-pre-treated OC [12, 13]. In a recent trial, paclitaxel was givenover 24 hours at 135 mg/m2 with carboplatin (AUC x 5),every 3 weeks, with considerable efficacy (90% ORR) incisplatin-sensitive patients, although the severe toxicityrequired G-CSF, and episodes of neutropenic sepsiswere frequent. Since paclitaxel-cisplatin is the standardfirst-line treatment [1], the oxaliplatin-paclitaxel combi-nation warrants consideration as a viable therapeuticoption for relapsing OC patients.

Our data support the effectiveness of oxaliplatin 130mg/m2 and paclitaxel 175 mg/m2 (three hours) adminis-tered every three weeks without growth factor support.Many agents show promising activity in pretreated OCpatients, but heamatological toxicity limits their useful-ness [14, 15]. The prevalence of platinum-sensitive recur-ring OC patients has increased, and they require active,

Downloaded from https://academic.oup.com/annonc/article-abstract/10/9/1125/173398by gueston 08 February 2018

1128

feasible and well-tolerated second-line regimens. Furtherclinical studies should be conducted to validate ourresults with pretreated OC patients, including taxane-exposed patients, before exploring its use in previouslyuntreated patients.

Acknowledgements

The authors thank the following physicians who treatedthe patients in the present cohort: Dr C. Borel (CentrePaul Strauss, Strasbourg), Dr H. Gautier (CentreMedico-chirurgical, Bligny), Dr D. Mayeur (HopitalAndre Mignot, Versailles), Dr J. L. Misset, Dr L. Bozec(Hopital Paul Brousse), Dr E. Pieter (Centre de Radio-therapie de Hartman, Neuilly sur Seine), and Dr Soulie(Centre Rene Huguenin, Saint-Cloud). The authorswould acknowledge M. Hughes for preparation of themanuscript.

References

1. McGuire WP, Hoskins WJ, Brady MF et al. Cyclophosphamideand cisplatin compared with pachtaxel and cisplatin in patientswith stage III and stage IV ovarian cancer. N Engl J Med 1996;334(1): 1-6.

2 Fink D, Nebel S, Norris PS et al. Enrichment for DNA mismatchrepair-deficient cells during treatment with cisplatin. Int J Cancer1998; 77 (5): 741-6.

3. Brown R, Hirsl GL, Gallagher WM et al. hMLHl expression andcellular responses of ovarian tumor cells to treatment with cyto-toxic anticancer agents. Oncogene 1997; 15 (I). 45-52.

4. Extra JM, Espie M, Calvo Fet al. Phase I Study of oxaliplatin inpatients with advanced cancer. Cancer Chemother Pharmacol1990; 25 (4): 299-303.

5. Chollet P, Bensmai'ne MA, Brienza S et al. Single-agent activityof oxaliplatin in heavily pretreated advanced epithelial ovariancancer. Ann Oncol 1996; 7 (10): 1065-70.

6. Soulie P, Bensmai'ne MA, Garrino C et al. Oxaliplatin/cisplatin(L-OHP/CDDP) combination in heavily pretreated ovarian can-cer. Eur J Cancer 1997; 33 (9): 1400-6.

7. Rowinsky EK. Update on the antitumor activity of pachtaxel in

clinical trials. Ann Pharmacother 1994; 28 (Suppl 5): S18-22-S32-34.

8 Skou G, Mangold G. Dexter D et al. Comparison of an oxali-platin-Taxol versus a carboplatin-Taxol regimens in the treatmentof the MV-522 human lung tumor xenograft. Proc Annu MeetAm Assoc Cancer Res 1996; 37: 375 (Abstr 2558)

9. Markman M. Hoskins W. Response to salvage chemotherapy inovarian cancer: A critical need for precise definitions of thetreated population. J Clin Oncol 1992, 10 (4): 513-4.

10. Reed E, Janik J, Bookman MA et al. High-dose carboplatin andrecombinant granulocyte-macrophage colony-stimulating factorin advanced-stage recurrent ovarian cancer J Clin Oncol 1993; 11(11): 2118-26.

11. Wils JA, van Geuns H, Wals J et al. Efficacy of carboplatin plusprimary prophylactic filgrastim (granulocyte colony stimulatingfactor) in relapsed ovarian cancer. A study of the GynecologicOncology Group of the Comprehensive Cancer Center Limburg.Anti-Cancer Drugs 1997; 8 (5): 432-5.

12. Pujade-Lauraine E, Guastalla JP, Weber B et al Efficacy andsafety of the combination paclitaxel-carboplatin in patients withpreviously treated advanced ovarian carcinoma: A multicenterFrench Groupe des Investigateurs Nationaux pour PEtude desCancers Ovariens phase II study. Semin Oncol 1997. 24 (5, Suppl15): S15-S35.

13 Rose PG, Fusco N, Fluellen L et al. Second-line therapy withpaclitaxel and carboplatin for recurrent disease following first-line therapy with paclitaxel and platinum in ovarian or peritonealcarcinoma. J Clin Oncol 1998; 16 (4): 1494-7.

14. Bertelsen K, Kaenn J. Simonsen E et al Is Cisplatin-Taxol (PT).the standard treatment in advanced ovarian cancer'' Eur J Cancer1998; 33 (Suppl 8): S138 (Abstr 614).

15. ICON2. Randomised trial of single-agent carboplatin againstthree-drug combination of CAP (cyclophosphamide. doxorubicin.and cisplatin) in women with ovarian cancer. ICON Collabora-tors. International Collaborative Ovarian Neoplasm Study. Lancet1998,352(9140): 1571-6.

Received 6 April 1999; accepted 16 June 1999.

Correspondence to.Dr J.-M. ExtraInstitut Curie26 rued'Ulm75005 PansFrance

Downloaded from https://academic.oup.com/annonc/article-abstract/10/9/1125/173398by gueston 08 February 2018