sign in sign up
<<
Home , Lapatinib

ANTICANCER RESEARCH 35: 5085-5090 (2015)

T-DM1 as a New Treatment Option for Patients with Metastatic HER2-positive in Clinical Practice

LAURA L. MICHEL1,2, JUSTO LORENZO BERMEJO3, ADAM GONDOS4, FREDERIK MARMÉ1,2 and ANDREAS SCHNEEWEISS1,2

1National Center for Tumor Diseases and 2Department of Obstetrics and Gynecology, University Hospital, Heidelberg, Germany; 3Institute of Medical Biometry and Informatics, University of Heidelberg, Germany; 4Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ), Heidelberg, Germany

Abstract. Aim: To compare results of - targeted approach (1, 2). Within the last few years, additional emtansine (T-DM1) treatment in our clinical practice with HER2-directed agents such as the monoclonal antibody data from phase III clinical trials. Patients and Methods: A and the dual epidermal receptor retrospective chart review of all 23 patients with metastatic (EGFR)/HER2 tyrosine inhibitor, lapatinib, were human epidermal 2 (HER2)-positive developed to overcome resistance to trastuzumab and provide breast cancer who were started on T-DM1 until April 2014 additional treatment options for these patients (3, 4). was performed. Results: Four patients (17.4%) received The newest agent, which was approved by the Food and T-DM1 as first-line, three (13.0%) as second-line, six Drug Administration (FDA) in February 2013 for HER2- (26.0%) as third-line, and 10 (43.5%) as fifth- or further-line positive is trastuzumab-emtansine therapy. Overall, the response rate (ORR) was 26.0%, (T-DM1, Roche, Basel). T-DM1 represents a new generation disease control rate 78.3% and median progression-free of cytotoxic drugs. It is an antibody–drug conjugate survival (PFS) 8.4 months. The only toxicities of grade 3 or consisting of the monoclonal antibody trastuzumab linked to more were (21.7%), thrombocytopenia (4.3%) and the cytotoxic agent emtansine (DM1), a microtubule elevation of enzymes (8.7%). ORR and PFS were polymerization inhibitor (5-7). T-DM1, therefore, combines similar to the TH3RESA and EMILIA trials. Compared to the the tumor selectivity of antibodies with the cytotoxic EMILIA study, we recorded higher rates of newly-diagnosed potential of chemotherapeutics. cerebral metastasis and cerebral progression in patients with T-DM1 binds to the HER2 receptor, is internalized and stable peripheral metastases. Conclusion: T-DM1 is effective undergoes lysosomal degradation. The active cytotoxic DM1 and well-tolerated even in intensively pre-treated patients. moiety is released inside HER2-expressing cells. This reduces systemic exposure. Additionally to inhibiting tubulin Over the past decade, the management of human epidermal polymerization similarly to the vinca alkaloids, T-DM1 also growth factor receptor 2 (HER2)-positive breast cancer has acts by trastuzumab-mediated HER2-targeting antitumor improved dramatically. The monoclonal antibody trastuzumab properties: it inhibits HER2 extracellular domain shedding, as was the first HER2-directed agent for the treatment of HER2- well as the phosphatidylinositol-3-kinase signaling pathway positive breast cancer. It was approved in 2000 in the EU. Its and induces antibody-dependent cellular cytotoxicity (7). introduction into clinical therapy represented a paradigm shift Results from multiple phase III studies have confirmed the in oncology from non-specific to a molecularly- efficacy and safety of T-DM1 in patients with metastatic breast cancer. The EMILIA (TDM4370) study is an international, randomized phase III study that compares the efficacy and safety of T-DM1 to plus lapatinib. A total of 991 Correspondence to: Laura L. Michel, MD, National Center for Tumor women with metastatic HER2-positive breast cancer whose Diseases, Heidelberg University Hospital, Im Neuenheimer Feld 460, disease progressed on or within 6 months after receiving D-69120 Heidelberg, Germany. Tel: +49 62215638795, e-mail: trastuzumab and who had received chemotherapy were [email protected] included (8). Patients received T-DM1 at a dose of 3.6 mg/kg Key Words: HER2-positive metastatic breast cancer, T-DM1, once every 21 days or oral lapatinib plus capecitabine. The chemotherapy. primary end-points were progression-free survival (PFS) and

0250-7005/2015 $2.00+.40 5085 ANTICANCER RESEARCH 35: 5085-5090 (2015) overall survival (OS). Patients had previously received a mg/kg. In cases of toxicities of grade 3 or more, the dose was reduced median of three systemic agents for metastatic disease. 12.0% by ~20%, corresponding to 3.0 mg/kg (level −1) and 2.4 mg/kg (level of the patients received T-DM1 as a first -line treatment. The −2). T-DM1 was given every three weeks. Echocardiography was performed every 12 weeks to rule out cardiotoxicity. median PFS was 9.6 months in the T-DM1 group vs. 6.4 For follow-up, patients were seen on a regular basis at least months in the lapatinib-plus-capecitabine group. The hazard quarterly at our Outpatient Clinic. According to our center's ratio (HR) for PFS was 0.65 with T-DM1 vs. capecitabine- guidelines, response was evaluated every third cycle by magnetic plus-lapatinib. The median survival was 30.9 months in the T- resonance imaging or computed tomographic scans and evaluation DM1-treated group vs. 25.1 months in the group treated with of tumor markers. lapatinib plus capecitabine (8). The TH3RESA study investigated the use of T-DM1 in Statistical analysis. Descriptive statistics were used to summarize patient and tumor characteristics. The overall response rate (ORR) comparison to the treatment of the physician's choice in was defined as complete and partial response (CR+PR) according to patients with HER2-positive breast cancer who had received the Response Evaluation Criteria in Solid Tumors (RECIST) version at least two prior regimens of HER2-directed therapy. The 1.1 (11). Disease control rate (DCR) was defined as CR plus PR plus primary end-points were PFS and OS. The median PFS was stable disease (SD) according to RECIST criteria. PFS was defined 6.2 months in the T-DM1-treated group vs. 3.3 months in the as the interval between the first dose of T-DM1 and the time of physician's choice group. Interim OS analysis showed a trend disease progression or death. OS was defined as the time interval favoring T-DM1 (stratified HR=0.552; 95% confidence between the first dose of T-DM1 and death. OS and PFS were examined using the Kaplan–Meier method. The follow-up of patients interval (CI)=0.369-0.826; p=0.0034) without crossing the without documented events at the last follow-up was censored. stopping boundary (9). Analysis of adverse events (AEs), combining available data Results from all single-agent T-DM1 studies to date including 884 T- DM1-exposed patients, revealed that the most commonly Patients' characteristics. Patient and tumor characteristics are reported all-grade AEs were fatigue (46.4%), (43.0%), summarized in Table I and reflected the typical population thrombocytopenia (32.2%), headache (29.4%), and of patients with MBC seen in our clinical practice. Overall, constipation (26.5%). The most common grade 3 to 4 AEs 23 patients with MBC were treated with T-DM1 between observed were thrombocytopenia (11.9%) and increased serum November 2010 to April 2014. The median follow-up was aspertate transaminase (AST) concentration (4.3%). There 285 (range=21-1016) days. At the cut-off date (October, 8th, were 12 AE-related deaths. AEs resulted in dose reductions in 2014), nine patients were still under therapy with T-DM1. 17.2% of patients and drug discontinuations in 7.0% (10). The median age was 54 (range=32-79) years. Overall 91.3% It is common belief that controlled clinical trials provide the of patients had visceral metastases. A total of 73.9% patients highest level of evidence. Yet it is also undisputed that clinical had received prior (neo-) adjuvant cytotoxic therapy; 91.3% trials are prone to selection bias and do not always reflect received prior anti-HER2 treatment either in the (neo-) clinical reality. After the introduction of T-DM1 in the EU at adjuvant (47.8%) or palliative (82.6%) setting, or both. Most the beginning of 2014, use of T-DM1 is still not common in patients (91.3%) had received prior antibody therapy with German clinical practice. At our Institution, T-DM1 has been trastuzumab for a median of 32 (range=5-120) months. integrated into routine treatment of HER2-positive metastatic Fourteen patients (60.9%) had been treated with lapatinib breast cancer (MBC). In this retrospective study, we report our and eight (34.8%) had received pertuzumab before. first experiences with T-DM1 in routine clinical practice and aim to determine how our treatment results compare to the Treatment. Details of T-DM1 treatment are outlined in Table data generated in controlled clinical trials. II. Four patients (17.4%) received T-DM1 as first-line treatment, 10 patients (43.5%) had received more than four Patients and Methods prior lines of systemic treatment for metastatic disease. All patients started with T-DM1 at 3.6 mg/kg. The median Patients. We performed a retrospective chart review of all 29 number of cycles administered was 10 (range=1-42). Dose patients who started treatment with T-DM1 at our Department from delay was required for 10 patients (43.5%) and dose November 2010 to April 2014. Six patients were excluded from our reduction was necessary for three patients (13.0%). patient cohort: two of them received T-DM1 in a post-neoadjuvant setting; in four other patients, therapy with T-DM1 was initiated at Efficacy. Efficacy data are summarized in Table III. No CR our Clinic and was continued after the first two cycles at other was observed. Overall tumor response was observed in six Institutions closer to patients’ homes. patients (26.0%). The DCR, defined as sum of CR, PR and Treatment details. Two patients of our study cohort were treated within SD, was 78.3%. Among the 23 women, five died (median a phase III study, receiving T-DM1 with/without pertuzumab. For all OS=11.4 months) and 15 women presented disease other patients, T-DM1 was started as monotherapy at a dose of 3.6 progression (median PFS time=6.6 months).

5086 Michel et al: T-DM1 in Routine Clinical Practice

Table I. Characteristics of patients treated with trastuzumab-emtansine at the National Center for Tumor Diseases in Heidelberg from November 2010 to April 2014.

Characteristic Patients, N (%)

Median age, years (range) 54 (32-79) Performance status 0 10 (43.5%) 1 9 (39.1%) 2 4 (17.4%) Estrogen receptor status Positive 13 (56.5%) Negative 10 (43.5%) Pattern of metastasis Visceral 21 (91.3%) Non-visceral 2 (8.7%) Organs with metastatic lesions Bone 11 (47.8%) Lung and pleura 7 (30.4%) Liver 11 (47.8%) Brain 6 (26.1%) Lymph nodes 12 (52.2%) Skin 3 (13.0%) Number of metastatic sites >3 3 (13.0%) 2-3 13 (56.5%) 1 7 (30.4%) Prior (neo-) adjuvant cytotoxic therapy Yes 17 (73.9%) No 6 (26.1%) Prior (neo-) adjuvant therapy with trastuzumab Yes 11 (47.8%) No 12 (52.2%) Prior systemic treatment lines Median (range) 2 (0-7) (cytotoxic/antibody) for recurrent 0 4 (17.4%) or metastatic disease 1 3 (13.0%) 2-3 6 (26.1%) ≥4 10 (43.5%) Prior HER2-directed therapy for metastatic disease Trastuzumab Average duration, months, (range) 32.7 (0-120) Yes 21 (91.3%) No 2 (8.7%) Lapatinib Average duration, months, (range) 8.6 (0-32) Yes 14 (60.9%) No 9 (39.1%) Pertuzumab Average duration, months, (range) 2.1 (0-10) Yes 8 (34.8%) No 15 (65.2%)

Figure 1 represents Kaplan-Meier curves for OS and PFS, (seven cases of progression) compared to 18% (standard with the number of patients at risk shown in the lower part of error: 15%) in women with 4-7 pre-treatments (eight cases the plots. The estimated one-year OS rate was 91% (standard of progressions, log-rank test probability value=0.02) (Figure error: 8%) in women with 0-3 pre-treatments (13 women and 1 C and D). two deaths), compared to 64% (23%) in women with 4-7 pre- treatments (10 women and three deaths). However, differences Toxicity. A total of 270 doses of T-DM1 were administered. in OS rates did not reach statistical significance (probability The incidence of toxicities grade 3 or more are summarized value from a log-rank test=0.16) (Figure 1 A and B). in Table IV. There was no grade 5 toxicity. Grade 3 The median PFS was 12.9 months (95% CI=5.7 to 31.5 thrombocytopenia was seen in one patient, mild months) in women with 0-3 pre-treatments, 6.6 months (95% thrombocytopenia (grade 1 and 2) was seen in four patients. CI=0.7 to 7.4 months) in women with 4-7 pre-treatments, Neutropenia grade 2 was seen in one patient. No case of and 8.4 months (95% CI=5.7 to 12.9 months) for the entire febrile neutropenia was documented. Elevation of liver investigated cohort. The one-year PFS rate was 53% enzymes of grade 3 and 4 were seen in two patients. Mild (standard error: 16%) in women with 0-3 pre-treatments elevation of liver enzymes (grade 1 and 2) was seen in four

5087 ANTICANCER RESEARCH 35: 5085-5090 (2015)

Figure 1. Kaplan–Meier curves for overall (A, B) and progression-free (C, D) survival, with the number of patients at risk shown at the bottom of the plots. A: Overall survival for the whole group of 23 patients that were treated with trastuzumab-emtansine. B: Overall survival according to the number of prior treatment lines (0-3 vs. 4-7). C: Progression-free survival for the overall population. D: Progression-free survival according to the number of prior treatment lines.

patients. The most frequent severe non-hematological toxicity years, respectively). A total of 17% of our patients had a was fatigue, observed in five patients (grade 3). Two patients performance status of 2, whereas only patients with developed mild skin changes (rash/eczema). One patient performance status of 0 and 1 were included in the EMILIA presented with stomatitis grade 3. No dose-limiting study. In all, 39% of patients in our cohort had a cardiotoxicity (grade 3, left ventricular ejection fraction performance status of 1 and 43% had a performance status decline or symptomatic congestive heart failure) was seen. In of 0 as compared to 39% and 60% of patients treated with summary, treatment termination because of side-effects was T-DM1 in the prospective clinical trial. Importantly, patients necessary in one patient only with grade 4 thrombocytopenia in our cohort had received more chemotherapy regimens for that was persistent even after dose reduction. A dose locally advanced or metastatic disease prior to study reduction was necessary in two additional patients because of inclusion: 60% in the prospective clinical trial had no or only stomatitis and decline in general condition. one prior chemotherapy regimens compared to only 30% of patients in our cohort (8). Discussion The ORR in the EMILIA trial was higher compared to our cohort (43.6% vs. 26.0%). Besides the low number of It is a challenge to compare results of clinical trials with patients, one explanation for this might be that patients in the those from routine clinical practice. First of all, patient EMILIA trial were less pre-treated and had received less populations will almost certainly differ. Compared to patients HER2-directed agents; pre-treatment with lapatinib was an in the EMILIA study, patients in our cohort were exclusion criterion. In our cohort, response rates were highly approximately the same age (median age=54 years vs. 53 dependent on the number of prior therapies (see Table III).

5088 Michel et al: T-DM1 in Routine Clinical Practice

Table II. Treatment details of our patient cohort.

Overall, N=23 First- and second-line, N=7 Third- and fourth-line, N=6 Fifth-line or more, N=10

Median number of cycles (range) 10 (1-45) 10 (2-45) 12.5 (5-21) 9 (1-22) Mean number of cycles 11.7 14.7 12.3 9.3 Patients with dose delays, N (%) 10 (43.5%) 4 (57.0%) 1 (16.7%) 5 (50.0%) Patients with dose reductions, N (%) 3 (13.0%) 1 (14.3%) 1 (16.7%) 1 (10.0%) Ended therapy because of side-effects 1 (4.3%) 0 0 1 (10.0%) Ended therapy because of PD/death 11 (47.8%) 5 (71.4%) 1 (16.7%) 5 (50.0%)

PD, Progressive disease.

Table III. Summary of efficacy data.

All, N=23 First-line, N=4 Second-line or Fourth-line or more, N=19 more, N=10

ORR (CR+PR) 6 (26.0%) 2 (50.0%) 4 (21.0%) 2 (20.0%) DCR (CR, PR and SD) 18 (78.3%) 4 (100%) 14 (73,7%) 7 (70.0%) PD 5 (21.7%) 0 (0%) 5 (26.3%) 3 (30.0%)

All, N=23 First to fourth Fourth-line or line N=13 more, N=10

Median PFS, (95% CI), months 8.4 (5.7 to 12.9) 12.9 (5.7 to 31.5) 6.6 (0.7 to 7.4) Estimated one-year OS rate (standard error) 0.81 (0.11) 0.91 (0.08) 0.64 (0.23) One-year PFS rate (standard error) 0.39 (1.12) 0.53 (0.16) 0.18 (0.15)

ORR, Overall response rate; DCR, disease control rate; SD, stable disease; PD, progressive disease; OS, overall survival; PFS, progression-free survival; CI confidence interval.

These facts might also explain the shorter median PFS in Table IV. Severe adverse events (grade 3/4) observed in patients treated our cohort (8.4 month vs. 9.6 months). The median PFS was with trastuzumab-emtansine. significantly higher in patients with 1-3 prior treatment lines Adverse event Grade 3, N (%) Grade 4, N (%) (cytotoxic/antibody, 12.9 months, 95% CI=5.7 to 31.5 months) (8). Non-hematological Patients in our cohort were more similar compared to Fatigue 5 (21.7%) 0 patients that received T-DM1 in the TH3RESA study; Infection 1 (4.3%)0 Hematological patients were approximately the same age (median age=54 Neutropenia 0 0 years vs. 53 years). Patients with a performance status of 2 Febrile neutropenia 0 0 were also included in the TH3RESA trial – even though Anemia 0 0 fewer compared to our patient cohort (17.4% vs. 5.0%). A Thrombopenia 1 (4.3%) 0 total of 33% of patients had received three or fewer prior Elevated liver enzymes 1 (4.3%) 1 (4.3%) chemotherapy regimens, 37% had received 4-5 prior N, Number of patients. chemotherapy regimens and 30% of patients had received more than five chemotherapy regimens for advanced breast cancer, compared to 57%, 26% and 17% of our patients (9). ORR in the TH3RESA trial was better comparable to our patients. Grade 3 fatigue was only observed in 2.4% of cohort (31% vs. 26%) than that in the EMILIA trial. The patients in the EMILIA study and in 2% of patients in the median PFS was lower in the TH3RESA trial compared to TH3RESA study. One explanation for the increased rates of our patients (6.2 month vs. 8.4 months) (9). fatigue seen in our cohort is that patients were more heavily Overall, T-DM1 was very well tolerated. Fatigue was the pre-treated compared to those of the EMILIA study and in most frequently reported grade 3 AE observed in 22% of inferior general condition. It is nearly impossible to separate

5089 ANTICANCER RESEARCH 35: 5085-5090 (2015) fatigue as a consequence of the current cytotoxic treatment 2 Ahmed S, Sami A and Xiang J: HER2-directed therapy: current from fatigue as a symptom of the underlying disease or treatment options for HER2-positive breast cancer. Breast previous treatments (8, 9). Cancer 22(2): 101-116, 2015. In the EMILIA trial, only patients with treated, 3 Slamon DJ, Leyland-Jones B, Shak S, Fuchs H, Paton V and Bajamonde A: Use of chemotherapy plus a monoclonal antibody asymptomatic cerebral metastases were included. A total of against HER2 for metastatic breast cancer that overexpresses 45 patients in the T-DM1 arm and 50 patients in the HER2. N Engl J Med 15;344(11): 783-792, 2001. capecitabin/lapatinib arm had CNS metastases at baseline. 4 Geyer CE, Forster J, Lindquist D, Chan S, Romieu CG, CNS progression was observed in 10/45 (22.2%) and 8/50 Pienkowski T, Jagiello-Gruszfeld A, Crown J, Chan A, Kaufman (16.0%) patients in the T-DM1 and capecitabin/lapatinib arm, B, Skarlos D, Campone M, Davidson N, Berger M, Oliva C, respectively. Primary diagnosis of CNS metastases occurred Rubin SD, Stein S and Cameron D: Lapatinib plus capecitabine in 9/450 (2.0%) patients in the T-DM1 arm and 3/446 (0.7%) for HER2-positive advanced breast cancer. N Engl J Med 28;355(26): 2733-2743, 2006. patients in the capecitabin/lapatinib arm (9, 12). 5 LoRusso PM, Weiss D, Guardino E, Girish S and Sliwkowski In our study six out of 23 (26%) patients had CNS MX: : a unique antibody-drug conjugate metastases at baseline. CNS progression was seen in two out in development for human receptor 2- of six patients (33%). Interestingly, in these two patients, positive cancer. Clin. Cancer Res 15;17(20): 6437-6447, 2011. isolated CNS progression occurred whereas the peripheral 6 Niculescu-Duvaz I: Trastuzumab emtansine, an antibody–drug metastases were stable. In these cases, treatment with T-DM1 conjugate for the treatment of HER2+ metastatic breast cancer. was continued in combination with whole-brain radiation. In Curr Opin Mol Ther 1;12(3): 350-360, 2010. 7 Junttila TT, Li G, Parsons K, Philips Lewis G and Sliwkowski four out of 23 (17.4%) patients, CNS metastases were MX: Trastuzumab-DM1 (T-DM1) retains all the mechanisms of primarily diagnosed, with three of them under therapy with T- action of trastuzumab and efficiently inhibits growth of lapatinib DM1 and one during treatment break due to side-effects. insensitive breast cancer. Breast Cancer Res Treat 21;128: 347- Three out of four patients had stable tumor masses outside the 356, 2010. CNS. In two patients, treatment with T-DM1 was continued 8 Verma S, Miles D, Gianni L, Krop IE, Welslau M, Baselga J, and additional whole-brain radiation was performed. At the Pegram M, Oh DY, Diéras V, Guardino E, Fang L, Lu MW, cut-off date, both patients had stable disease. Olsen S, Blackwell K and EMILIA Study Group: Trastuzumab The risk of CNS progression under therapy, as well as emtansine for HER2-positive advanced breast cancer. N Engl J Med 8;367(19): 1783-1791, 2012. the number of primary diagnosed CNS metastases, was 9 Krop IE, Kim SB, González-Martín A, LoRusso PM, Ferrero significantly higher in our cohort compared to patients in JM, Smitt M, Yu R, Leung AC, Wildiers H and TH3RESA study the EMILIA trial (33.3% vs. 22.2% and 17.4% vs. 2.0%, collaborators: Trastuzumab emtansine versus treatment of respectively) (12). One explanation for this observation physician's choice for pretreated HER2-positive advanced breast might be that our patients had more advanced tumor cancer (TH3RESA): a randomised, open-label, phase III trial. disease so that their general risk of developing CNS Lancet Oncol 15(7): 689-699, 2014. metastases was higher. This hypothesis is supported by the 10 Diéras V, Harbeck N, Budd GT, Greenson JK, Guardino AE, Samant M, Chernyukhin N, Smitt MC and Krop IE: number of patients with CNS metastases at baseline: 9% Trastuzumab emtansine in human epidermal growth factor in the T-DM1 arm of the EMILIA trial vs. 26% of patients receptor 2-positive metastatic breast cancer: an integrated safety in our cohort. analysis. J Clin Oncol 1;32(25): 2750-2757, 2014. In summary, T-DM1 was an effective and well-tolerated 11 Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent therapy in routine clinical practice that offers a new D, Ford R, Dancey J, Arbuck S, Gwyther S, Mooney M, therapeutic option, even for heavily pre-treated patients with Rubinstein L, Shankar L, Dodd L, Kaplan R, Lacombe D and HER2-positive breast cancer. PFS and OS were highly Verweij J: New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 45(2): dependent on the number pretreatments and were comparable 228-247, 2009. to results of the EMILIA and TH3RESA study. We observed 12 Krop IE, Lin NU, Blackwell K, Guardino E, Huober J, Lu M, a higher incidence of newly diagnosed cerebral metastases Miles D, Samant M, Welslau M and Diéras V: Trastuzumab and isolated cerebral progress under therapy with T-DM1. emtansine (T-DM1) versus lapatinib plus capecitabine in patients Further studies are necessary to further examine this finding with HER2-positive metastatic breast cancer and central nervous in larger patient cohorts. system metastases: a retrospective, exploratory analysis in EMILIA. Ann Oncol 26(1): 113-119, 2015. References

1 Molina MA, Codony-Servat J, Albanell J, Rojo F, Arribas J and Baselga J: Trastuzumab (herceptin), a humanized anti-HER2 receptor monoclonal antibody, inhibits basal and activated HER2 Received May 8, 2015 ectodomain cleavage in breast cancer cells. Cancer Res 61(12): Revised June 17, 2015 4744-4749, 2001. Accepted June 19, 2015

5090