P h a s e I I T r i a l o f Z D 1 8 3 9 i n R e c u r r e n t o r M e t a s t a t i c S q u a m o u s
`C e l l C a r c i n o m a o f t h e H e a d a n d N e c k
`
`By Ezra E. W. Cohen, Fred Rosen, Walter M. Stadler, Wendy Recant, Kerstin Stenson, Dezheng Huo, and Everett E. Vokes
`
`Purpose: The epidermal growth factor receptor (EGFR) is
`a mediator of squamous cell carcinoma of the head and
`neck (SCCHN) development. ZD1839 is an orally active,
`selective EGFR tyrosine kinase inhibitor. This phase II study
`sought to explore the activity, toxicity, and pharmacody-
`namics of ZD1839 in SCCHN.
`Patients and Methods: Patients with recurrent or meta-
`static SCCHN were enrolled through the University of Chi-
`cago Phase II Consortium. Patients were allowed no more
`than one prior therapy for recurrent or metastatic disease
`and were treated with single-agent ZD1839 500 mg/d.
`Patient tumor biopsies were obtained and stained immuno-
`histochemically for EGFR, extracellular signal-regulated ki-
`nase 1 (ERK1), and phosphorylated ERK1 (p-ERK). Study end
`points included response rate, time to progression, median
`survival, and inhibition of p-ERK.
`Results: Fifty-two patients were enrolled (40 male and
`12 female) with a median age of 59 years (range, 34 to 84
`
`years). Fourteen patients received ZD1839 through a feed-
`ing tube. Half the cohort received ZD1839 as second-line
`therapy. Forty-seven patients were assessable for re-
`sponse, with an observed response rate of 10.6% and a
`disease control rate of 53%. Median time to progression and
`survival were 3.4 and 8.1 months, respectively. The only
`grade 3 toxicity encountered was diarrhea in three patients.
`Performance status and development of skin toxicity were
`found to be strong predictors of response, progression, and
`survival. Ten biopsy samples were assessable and revealed
`no significant change in EGFR or p-ERK expression with
`ZD1839 therapy.
`Conclusion: ZD1839 has single-agent activity and is well
`tolerated in refractory SCCHN. In contrast to other reports,
`development of skin toxicity was a statistically significant
`predictor of response and improved outcome.
`J Clin Oncol 21:1980-1987. © 2003 by American
`Society of Clinical Oncology.
`
`SQUAMOUS CELL carcinoma of the head and neck (SCCHN)
`
`often presents as a locally advanced disease; however, more
`than 50% of patients will eventually develop incurable local or
`metastatic disease. For these patients, therapeutic options are often
`palliative, while systemic chemotherapy has yet to demonstrate a
`substantial improvement in survival and produces considerable
`toxicity. Phase III randomized trials in patients with recurrent or
`metastatic SCCHN have demonstrated single-agent response rates
`between 10% and 15% and median survivals of 6 to 8 months, even
`with the use of combination chemotherapy.1-6
`
`From the Sections of Hematology/Oncology, Otolaryngology-Head and
`Neck Surgery, and Urology, and Departments of Medicine, Pathology,
`Surgery, Health Studies, and Radiation and Cellular Oncology, University of
`Chicago; and Section of Hematology/Oncology, Department of Medicine,
`University of Illinois-Chicago, Chicago, IL.
`Submitted October 7, 2002; accepted February 26, 2003.
`Supported in whole or in part by Federal funds from the National Cancer
`Institute, National Institutes of Health, Bethesda, MD, under Contract No.
`N01-CM-17102; the University of Chicago Cancer Research Center, Chi-
`cago, IL (grant no. P30 CA14599); and a private donation from The Francis
`Lederer Foundation, Chicago, IL.
`Presented in part at the Second International Chicago Symposium on
`Cancers of the Chest, Head, and Neck, Chicago, IL, October 4-6, 2001, and
`the Annual Meeting of the American Society of Clinical Oncology, May
`18-21, 2002.
`Address reprint requests to Ezra E.W. Cohen, MD, University of Chicago,
`5841 S Maryland Ave, MC 2115, Chicago,
`IL 60637-1470; email:
`ecohen@medicine.bsd.uchicago.edu.
`© 2003 by American Society of Clinical Oncology.
`0732-183X/03/2110-1980/$20.00
`
`Since the first description of the epidermal growth factor
`receptor (EGFR) in 1980,7 interest has grown in targeting this
`protein in cancer therapy. Expression of EGFR has been linked
`to carcinogenesis, metastasis, and survival in SCCHN patients.8
`Phosphorylation of EGFR cytoplasmic tyrosine residues initiates
`a cascade of signals that includes activation of the mitogen-
`activated protein kinase pathway.8 The mitogen-activated pro-
`tein kinase pathway culminates in activation and nuclear trans-
`location of the extracellular signal-regulated kinase (ERK) 1 and
`2 and transcription of its target genes.9 Preclinical studies have
`confirmed that interruption of EGFR phosphorylation can inhibit
`these downstream activation events, lead to cell cycle arrest, and
`compromise tumor growth.10-12
`ZD1839 (gefitinib) is an oral, low-molecular-weight anilino-
`quinazoline that reversibly inhibits EGFR tyrosine kinase activ-
`ity. It has demonstrated an acceptable toxicity profile in phase I
`trials with predictable pharmacokinetics that established dose,
`schedule, and dose-limiting toxicity.13
`This phase II trial was undertaken to assess the activity and
`tolerability of ZD1839 in recurrent or metastatic SCCHN given
`either orally or via gastrostomy tube at a fixed dose of 500 mg/d.
`In addition, this study sought to delineate the pharmacodynamics
`of ZD1839 in tumor tissue before and after therapy by examining
`biopsy specimens by immunohistochemistry for EGFR, ERK,
`and their phosphorylated forms.
`
`PATIENTS AND METHODS
`
`Eligibility
`
`This study enrolled patients with recurrent or metastatic SCCHN who
`were considered ineligible for curative surgery or radiotherapy. Patients were
`
`1980
`
`Journal of Clinical Oncology, Vol 21, No 10 (May 15), 2003: pp 1980-1987
`DOI: 10.1200/JCO.2003.10.051
`
`Downloaded from ascopubs.org by University of Chicago Library, Dr. Mark Ratain on July 27, 2017 from 205.208.061.064
`
`Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
`
`Ex. 1078-0001
`
`

`

`ZD1839 FOR SQUAMOUS CELL CARCINOMA
`
`enrolled at selected centers participating in the University of Chicago Phase
`II Consortium. Patients were required to have measurable disease as defined
`by Response Evaluation Criteria in Solid Tumors; were not allowed any prior
`EGFR-based therapy; were allowed no more than one prior systemic therapy
`for incurable, recurrent, or metastatic disease; and were allowed no chemo-
`therapy or radiotherapy within 4 weeks of study entry.
`Patients had to be at least 18 years of age, nonpregnant, have a life
`expectancy of 3 months or more, and an Eastern Cooperative Oncology
`Group performance status of 2 or less. Normal organ and marrow function were
`necessary and were defined as a leukocyte count ⱖ 3,000/␮L, an absolute
`neutrophil count ⱖ 1,500/␮L, a platelet count ⱖ 100,000/␮L, a total bilirubin
`within normal institutional limits, plasma AST and ALT levels ⱕ 2.5 times the
`institutional upper limit of normal, and a creatinine level ⱕ 1.5 mg/dL.
`All patients were required to understand and sign the applicable institu-
`tional review board’s approved informed consent document.
`
`Treatment Plan and Dose Modifications
`
`ZD1839 was administered to all patients at a fixed continuous dose of 500
`mg/d. Patients unable to swallow tablets were allowed to dissolve ZD1839 in
`water. All patients were given baseline ophthalmologic assessments, which
`included visual acuity and slit-lamp examinations.
`illness
`intercurrent
`Therapy was continued until disease progression,
`preventing further administration, unacceptable toxicity, or patient decision.
`Toxicity was graded using the National Cancer Institute common toxicity
`criteria version 2.0.
`Patients who experienced grade 2 skin rash, nausea, or diarrhea that was
`unacceptable had therapy temporarily held until resolution to grade 1 or less.
`If, on restarting therapy, the toxicity continued, the dose was lowered to 250
`mg. Other grade 2 nonhematologic toxicities required dose reduction to 250
`mg. Any grade 3 or 4 toxicity required temporary discontinuation of therapy
`until resolution to grade 1 or less and reinstitution at 250 mg. Patients whose
`toxicity did not resolve after 2 weeks of discontinuation or who required a
`second dose reduction were removed from study. Once a patient’s dose was
`reduced, it was not subsequently increased.
`
`Response Assessment
`
`Patients were re-evaluated clinically at least every 4 weeks and radio-
`graphically every 8 weeks. The same evaluation modality was used through-
`out the study. Response guidelines as defined by Response Evaluation
`Criteria in Solid Tumors were used,14 defining all responses after at least 8
`weeks of therapy as either a complete response (CR), a partial response (PR),
`progressive disease (PD), or stable disease (SD). We defined disease control
`as the sum of patients achieving a CR, PR, or SD. Confirmation of all
`responses was required after 4 weeks. The National Cancer Institute’s
`Clinical Trials Monitoring Branch independently reviewed all patients who
`responded, had tumor shrinkage, or had prolonged stable disease.
`
`Biopsy and Tissue Preparation
`
`Patients who had accessible tissue were randomly assigned to undergo
`biopsy either before therapy (pre) or at 7 weeks of therapy (post). Biopsies
`were performed with 1% xylocaine on an outpatient basis using a 14-guage
`biopsy needle. Tissue was instantly placed in Tissue Freezing Medium
`(Triangle Biomedical Sciences, Durham, NC) and 2-methylbutane in liquid
`nitrogen and stored at ⫺80°C. The study biopsies were stained for EGFR,
`ERK, and phosphorylated ERK (p-ERK)–tyrosine residue 204 on ERK-1.
`
`Immunohistochemistry
`
`The frozen tissues were sectioned into 6-␮m slices and fixed in 4%
`paraformaldehyde for 10 minutes. After the slides were rinsed, they were
`incubated in 3% hydrogen peroxide for 5 minutes and then 10% normal goat
`serum in 0.025% Triton X-100 phosphate-buffered saline for 20 minutes.
`The slides were incubated with either ERK-1 (1 ␮g/mL; Santa Cruz
`Biotechnology, Santa Cruz, CA), p-ERK (8 ␮g/mL; Santa Cruz Biotechnol-
`ogy), or EGFR antibody (1:25; Cell Signaling Technology, Beverly, MA) for
`1 hour at room temperature in a humidity chamber. After slides were washed
`
`1981
`
`in phosphate-buffered saline, they were incubated with EnVision Systems
`(DAKO A/S, Glostrup, Denmark) antimouse or antirabbit kit for 30 minutes
`at room temperature. The antigen-antibody binding was detected by 3,3⬘-
`diaminobenzidine chromogen system (DAKO A/S). The slides were briefly
`immersed in hematoxylin for counterstaining and evaluated by light micros-
`copy. ERK and p-ERK negative controls used both peptide absorption
`blocking and isotype-specific immunoglobulin. Isotype-specific immuno-
`globulin was used as a negative control for EGFR.
`Samples that were adequate were evaluated further using a 4-point scoring
`system on the basis of the number of cells that stained positively (0 ⫽ no
`staining; 1⫹ ⫽ ⬍10%; 2⫹ ⫽10% to 50%; 3⫹ ⫽ ⬎50%). Histologic
`examination was performed on all samples by a single pathologist (W.R.)
`who was blinded to timing of biopsy and response data.
`
`Statistical Analysis
`
`The trial used a two-stage design requiring the enrollment of 22 patients
`onto the first stage and an additional 24 patients onto the second stage. If at
`the end of the first stage fewer than two responses were observed and more
`than 14 patients experienced disease progression within 2 months, the trial
`would be stopped. Otherwise, if more than five responses were observed or
`fewer than 29 patients experienced disease progression within 2 months
`among the total 46 patients, this would be sufficient to reject the null
`hypothesis and conclude that ZD1839 warrants further study. This design
`provided an alpha level of 10% and a power between 0.74 and 0.90,
`depending on different alternative scenarios.
`The primary end points were response rate and time to progression (TTP).
`Secondary end points included survival, toxicity, and correlations of staining
`with response. All patients who met eligibility criteria and were assessable
`for response were included in the efficacy analysis. All patients who were
`registered and received drug were included in the toxicity analysis. Data
`were updated to June 24, 2002.
`TTP and survival were measured from date of registration until disease
`progression or death, respectively, and were summarized by Kaplan-Meier
`curves. Factors related to response or lack of early progression were analyzed
`using the Fisher’s exact test, and factors related to survival were analyzed
`using the log-rank test and Cox proportional hazards model. Staining
`intensity comparison was performed using a Wilcoxon rank sum test. The
`correlation between the staining level and tumor response was evaluated
`using a Wilcoxon rank sum test. All statistical analyses were conducted at the
`.05 level of significance.
`
`RESULTS
`
`Patients and Eligibility
`
`Fifty-two patients were enrolled from March to October 2001.
`Their characteristics are listed in Table 1. Five patients were
`registered but were not assessable for response for the following
`reasons: one patient had a serum creatinine level greater than the
`eligibility limit (this patient never received the drug), one patient
`died of an unknown cause during cycle 1, two patients on further
`review did not have head and neck cancer (one patient had
`non–small-cell lung cancer and one patient had benign disease),
`and one patient was removed from study because of possible
`toxicity (transverse myelopathy) during cycle 1. The latter three
`patients are included in the toxicity analysis.
`Prior therapy administered to the 47 assessable patients is
`listed in Table 1. Prior chemotherapy was administered as either
`part of an initial curative intent chemoradiotherapy regimen (n ⫽
`30) or as palliative treatment of incurable disease (n ⫽ 23). Of
`the 40 patients (85%) who received chemotherapy at any time
`during their treatment, 28 (70%) had a prior platinum-containing
`regimen. Half the patients (49%) had experienced treatment
`
`Downloaded from ascopubs.org by University of Chicago Library, Dr. Mark Ratain on July 27, 2017 from 205.208.061.064
`
`Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
`
`Ex. 1078-0002
`
`

`

`1982
`
`Table 1. Patient Demographics
`
`Characteristic
`
`Total
`Male
`Female
`Age, years
`Median
`Range
`Disease status at enrollment
`Locally recurrent
`Metastatic*
`ECOG performance status†
`0
`1
`2
`Prior therapy (n ⫽ 47)
`Surgery
`Radiotherapy
`With chemotherapy
`Alone
`Chemotherapy
`As part of the radiotherapy
`As part of radiotherapy then for recurrent/
`metastatic disease§
`Only for recurrent or metastatic disease§
`
`%
`
`100
`77
`23
`
`Patients
`
`59
`34-84
`
`44
`56
`
`21
`62
`17
`
`89
`96
`67‡
`33‡
`85
`43‡
`33‡
`
`25‡
`
`No.
`
`52
`40
`12
`
`23
`29
`
`10
`29
`8
`
`42
`45
`30
`15
`40
`17
`13
`
`10
`
`Abbreviation: ECOG, Eastern Cooperative Oncology Group.
`*These patients may also have had local recurrence simultaneously with
`metastatic disease.
`†This is reported for the 47 assessable patients.
`‡These data reflect percentages of the subgroup, ie, radiotherapy or
`chemotherapy.
`total assessable) received a prior
`§Twenty-three patients (49% of
`regimen for recurrent/metastatic disease.
`
`COHEN ET AL
`
`Fig 1. Kaplan-Meier curve of (A) time to progression and (B) overall survival.
`
`failure with a prior systemic regimen for incurable recurrent or
`metastatic disease and therefore received ZD1839 as second-line
`palliative therapy. The median time from completing prior
`therapy to registration was 4 months (range, 1 to 78.8 months).
`The protocol allowed administration of ZD1839 via feeding
`tube. In total, 14 patients received ZD1839 via this route. This
`group, albeit small, did not exhibit any clinical or statistical
`differences with respect to response, toxicity, or survival com-
`pared with patients who took ZD1839 orally.
`Follow-up for patients continued after disease progression
`until death. In total, 14 patients received subsequent therapy
`consisting of chemoradiotherapy in three patients and systemic
`chemotherapy in 11 patients.
`
`Treatment Responses
`
`Two patients had either CR or PR and nine patients experi-
`enced disease progression within 2 months among the 22
`patients entered during the first stage; five patients had either CR
`or PR and 22 patients experienced disease progression within 2
`months among the first 46 assessable patients. Thus from this
`standpoint, the drug can be declared sufficiently active to warrant
`further study. Of the total 47 assessable patients, we observed
`one CR and four PRs for an overall response rate of 10.6% (95%
`confidence interval [CI], 3.5% to 23.1%). A total of 20 patients
`(42.6%) had SD (95% CI, 28.3% to 57.8%) as their best
`
`response, including five patients experiencing minor responses
`that did not meet criteria for PR. Therefore, as defined above,
`53% of the patients experienced some degree of disease control.
`The remaining 22 patients (46.8%) had progressive disease at
`initial re-evaluation (95% CI, 32.1% to 61.9%).
`Of the responding patients,
`three had metastatic disease
`(visceral or soft tissue) and two had local recurrences. At last
`update, four of the five responders had experienced disease
`progression, with a median duration of response of 1.6 months
`(range, 1.2 to 11 months).
`
`TTP and Survival
`
`In total, 45 patients have eventually developed progressive
`disease, with two patients remaining on study (Fig 1A). The
`median TTP was 3.4 months (95% CI, 1.8 to 3.6 months). By 3,
`6, and 9 months, 53.2%, 12.8%, and 6.4% of patients, respec-
`tively, had not experienced disease progression.
`With a median follow-up time of 11.4 months, median
`survival has reached 8.1 months (95% CI, 5.2 to 9.4 months) for
`the entire cohort, with a 1-year survival probability of 29.2%
`(Fig 1B). Of the 47 assessable patients, 14 are still alive.
`
`Toxicity
`
`Fifty patients were included in the toxicity analysis. One
`patient never received ZD1839 and one patient died during cycle
`
`Downloaded from ascopubs.org by University of Chicago Library, Dr. Mark Ratain on July 27, 2017 from 205.208.061.064
`
`Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
`
`Ex. 1078-0003
`
`

`

`ZD1839 FOR SQUAMOUS CELL CARCINOMA
`
`1983
`
`Table 2.
`
`Toxicity Observed by Grade
`
`Grade
`
`Table 3.
`
`Factors Related to Response
`
`CR/PR/SD
`
`PD
`
`Toxicity
`
`%* of Patients
`
`Skin
`Keratitis
`Anorexia
`Nausea
`Vomiting
`Diarrhea
`ALP
`AST
`ALT
`Hypercalcemia
`Creatinine
`Dyspnea
`
`48
`4
`26
`18
`12
`50
`4
`12
`4
`20
`2
`
`1
`
`15
`0
`8
`5
`2
`18
`1
`5
`2
`5
`0
`1
`
`2
`
`9
`2
`2
`2
`4
`4
`1
`1
`0
`2
`1
`2
`
`3
`
`0
`0
`3
`2
`0
`3
`0
`0
`0
`2
`0
`0
`
`4
`
`0
`0
`0
`0
`0
`0
`0
`0
`0
`1
`0
`0
`
`Abbreviation: ALP, alkaline phosphatase.
`*Fifty patients were assessable for toxicity.
`
`1 without toxicity data available. Toxicities encountered are
`listed in Table 2. The most common toxicities observed were
`dermatologic and gastrointestinal. The integumentary toxicity
`included an acneiform skin rash, brittle hair, and onycholysis. Of
`the 24 patients who developed skin toxicity, 20 did so in cycle 1
`and an additional two patients did so by cycle 2. The only patient
`to discontinue therapy because of toxicity did so by choice
`because of intolerable acneiform rash (grade 2). As listed in
`Table 2, the most common gastrointestinal toxicity was diarrhea,
`which required dose reduction in four patients (three patients
`with grade 3 and one patient with grade 2 diarrhea). Subsequent
`to dose modification, all patients were able to continue therapy at
`the lower dose.
`Some adverse events encountered during the trial were possibly
`related to the agent or the disease. One patient experienced cervical
`myelopathy with urinary and stool incontinence 7 days after starting
`ZD1839. These symptoms abated and completely resolved within 4
`weeks of discontinuing therapy. Magnetic resonance imaging of the
`spinal cord at the time was nondiagnostic.
`Three patients developed cellulitis at sites of active skin
`involvement during therapy. The findings consisted of marked
`inflammation with warmth and erythema. Although an infectious
`diagnosis was made, cultures were sterile in all cases. In
`addition, three patients experienced hemorrhages at disease sites
`while receiving therapy. One of these events was fatal, whereas
`another required transfusion of two units of packed RBCs. Both
`of these patients had tumor shrinkage radiographically.
`As shown in Table 2, 20% of patients on study experienced
`some degree of hypercalcemia. None of these patients had
`evidence of bone metastasis. In all but one of these patients (a
`patient with a transient grade 1 value), the finding of hypercal-
`cemia preceded radiographic evidence of nonskeletal disease
`progression at their next evaluation.
`
`Factors Related to Response, Progression, and Survival
`
`Additional analysis of factors related to disease control re-
`vealed that only performance status and development of skin
`toxicity were predictive (Table 3) Prior therapy of any kind,
`duration from prior therapy, or administration of ZD1839 as
`
`Factor
`
`No.
`
`Performance status
`0
`1
`2
`ZD1839 as
`Second-line therapy
`First-line therapy
`Prior chemoradiotherapy
`Yes
`No
`Any chemotherapy
`Yes
`No
`Skin toxicity
`Yes
`No
`Diarrhea
`Yes
`No
`
`8
`17
`0
`
`12
`13
`
`16
`9
`
`22
`3
`
`17
`8
`
`17
`8
`
`%
`
`32
`68
`
`48
`52
`
`64
`36
`
`88
`12
`
`68
`32
`
`68
`32
`
`No.
`
`2
`12
`8
`
`11
`11
`
`14
`8
`
`18
`4
`
`5
`16
`
`8
`13
`
`P*
`
`.001
`
`.99
`
`.99
`
`.69
`
`.004
`
`.074
`
`%
`
`9
`55
`36
`
`50
`50
`
`64
`36
`
`82
`18
`
`24
`76
`
`38
`62
`
`Abbreviations: CR, complete response; PR, partial response; SD, stable
`disease; PD, progressive disease.
`*Fisher’s exact test.
`
`first- or second-line therapy did not predict for response or
`disease control.
`Factors related to progression are shown in Table 4. Similar to
`the response analysis, baseline performance status was strongly
`associated with TTP (P ⬍ .0001). Because development of skin
`toxicity was closely linked with disease control, this toxicity also
`predicted longer progression-free survival (4.3 v 2.1 months;
`P ⫽ .0002). In addition, patients who enrolled with metastatic
`disease experienced disease progression more rapidly than did
`those with locally recurrent disease (2.8 v 4.1 months; P ⫽ .03).
`A number of factors were associated with favorable survival,
`including baseline performance status (P ⬍ .0001), achieving a
`response (P ⫽ .0002) or disease control (P ⫽ .0001), and
`development of skin toxicity (Table 5 and Fig 2). Interestingly,
`patients who developed skin toxicity had a greater than two-fold
`median survival compared with patients who did not (11.1 v 5.3
`months; P ⫽ .001). The only other toxicity encountered with
`frequency (diarrhea) did not predict survival (P ⫽ .12). There
`was a nonsignificant trend toward improved survival in patients
`receiving ZD1839 as first-line therapy (P ⫽ .25).
`A multiple Cox proportional hazards model was used to
`identify independent prognostic factors of survival. In the
`unadjusted analysis, development of skin rash carried a hazard
`ratio of death of 0.30 (95% CI, 0.14 to 0.65). When performance
`status was adjusted for, patients with skin toxicity had a longer
`survival (P ⫽ .046), with a hazard ratio of 0.43 (95% CI, 0.19 to
`0.98). However, after adjustment for disease control, skin rash
`was not independently predictive of survival, with a hazard ratio
`of 0.49 (95% CI, 0.21 to 1.18), likely because of a strong
`correlation of skin rash with disease control. Conversely, disease
`control predicted prolonged survival when performance status or
`
`Downloaded from ascopubs.org by University of Chicago Library, Dr. Mark Ratain on July 27, 2017 from 205.208.061.064
`
`Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
`
`Ex. 1078-0004
`
`

`

`1984
`
`COHEN ET AL
`
`Table 4.
`
`Factors Related to Time to Progression
`
`Factor
`
`No. of Patients
`
`Median TTP
`(months)
`
`95% CI
`
`3 Months*
`(%)
`
`6 Months*
`(%)
`
`P†
`
`Performance status
`0
`1
`2
`Disease status
`Metastasis
`Recurrent
`ZD1839 as
`First-line therapy
`Second-line therapy
`Any prior chemotherapy
`Yes
`No
`Skin toxicity
`Yes
`No
`Diarrhea
`Yes
`No
`
`10
`29
`8
`
`28
`19
`
`23
`24
`
`40
`7
`
`22
`24
`
`25
`21
`
`5.4
`3.4
`1.6
`
`2.8
`4.1
`
`3.4
`3.6
`
`3.4
`1.4
`
`4.3
`2.1
`
`3.6
`1.8
`
`1.6 to 9.0
`1.8 to 3.6
`0.6 to 2.1
`
`1.6 to 3.5
`2.1 to 5.4
`
`1.7 to 3.5
`1.6 to 5.4
`
`2.1 to 3.9
`0.6 to 3.7
`
`3.1 to 5.5
`1.6 to 3.4
`
`2.8 to 5.3
`1.6 to 3.4
`
`80
`59
`0
`
`46
`63
`
`57
`50
`
`58
`29
`
`77
`33
`
`68
`38
`
`40
`7
`0
`
`7
`21
`
`4
`21
`
`13
`14
`
`23
`4
`
`16
`10
`
`⬍ .0001
`
`.03
`
`.35
`
`.15
`
`.0002
`
`.12
`
`Abbreviations: TTP, time to progression; CI, confidence interval.
`*Three- and 6-month columns represent percentage of patients who did not experience disease progression.
`†P value for log-rank test.
`
`skin toxicity were adjusted for (hazard ratio ⫽ 0.40 [95% CI,
`0.17 to 0.91] or 0.38 [95% CI, 0.16 to 0.88], respectively).
`
`Pharmacodynamic Studies
`
`A total of 14 samples were collected, 10 of which were
`adequate for interpretation: six before (pre) and four at 7 to 8
`weeks of therapy (post). The results of immunohistochemical
`staining are tabulated in Table 6. There was no difference
`statistically in EGFR (P ⫽ .13) staining intensity between the
`pre- and posttherapy samples. However, ERK staining intensity
`was statistically higher in the posttherapy samples (P ⫽ .02).
`Despite the higher intensity of ERK staining in the posttherapy
`samples, a consequent increase in p-ERK staining intensity was
`not observed (P ⫽ .90 for p-ERK pre v post). A correlation
`between response and staining was not observed for any of the
`proteins, although the value for patients who stained lower for
`p-ERK did approach significance (P ⫽ .11).
`
`DISCUSSION
`This is the first clinical trial to use ZD1839, a small-molecule
`tyrosine kinase inhibitor (TKI) of EGFR, in SCCHN. With 47
`assessable patients, the study was able to demonstrate activity
`and tolerability of this agent in patients with incurable disease.
`Although the overall response rate was modest (10.6%), fewer
`than half of the cohort experienced disease progression at first
`evaluation, with favorable TTP (3.4 months) and median sur-
`vival (8.1 months). These results would compare reasonably well
`with reported data of single-agent or combination chemotherapy
`regimens in phase III trials, with the added benefit of less
`toxicity.1-6,15-17
`Other EGFR inhibitor trials in SCCHN have shown remark-
`ably similar results. Cetuximab, a monoclonal antibody directed
`
`at the EGFR, yielded response rates of 11% when combined with
`platinum therapy in two separate phase II trials in platinum-
`refractory patients18,19 and 23% when combined with cisplatin as
`first-line therapy.20 Another small-molecule TKI, OSI-774, given
`to patients similar to those in our study produced a response rate of
`6%.21 The acneiform skin rash reported here has been consistently
`observed in all trials of EGFR inhibitors.13,22-28
`The acneiform skin rash is of special interest because it likely
`represents a toxicity that is inherently related to these agents’
`mechanism of action. Other reports have noted a correlation
`between the presence of the rash and response—an association
`that has not retained statistical significance when rash is related
`to TTP or survival.18,19,29 Nevertheless, in this study, develop-
`ment of rash was associated with statistically and clinically
`meaningful improvements in TTP and overall survival, likely
`related to the strong correlation observed between rash and
`disease control. Notably, however, two large monotherapy trials
`in non–small-cell
`lung cancer failed to show a correlation
`between response and skin rash.30,31
`These differences between the current study and prior reports
`could stem from several factors, including the disease studied or
`the scoring of toxicities. The great majority of patients who
`developed the rash did so before their first disease re-evaluation
`at 8 weeks; it is, therefore, unlikely that the association is related
`to patients being more likely to develop rash the longer they
`remain on therapy. However, SCCHN is a disease that almost
`universally expresses EGFR. In addition, the epithelium of the
`upper aerodigestive tract is closely related to skin both structur-
`ally and functionally more so than other mucosal surfaces of
`gastrointestinal, respiratory, or glandular tissues. Moreover, the
`investigators in this study counted all integumentary toxicity,
`including hair and nail, and asked patients to undress for a full
`
`Downloaded from ascopubs.org by University of Chicago Library, Dr. Mark Ratain on July 27, 2017 from 205.208.061.064
`
`Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
`
`Ex. 1078-0005
`
`

`

`ZD1839 FOR SQUAMOUS CELL CARCINOMA
`
`1985
`
`Table 5.
`
`Factors Related to Overall Survival
`
`Factor
`
`N*
`
`Median Survival
`(months)
`
`95% CI
`
`6-Month Survival
`(%)
`
`1-Year Survival
`(%)
`
`P†
`
`Performance status
`0
`1
`2
`Disease status
`Metastasis
`Recurrent
`ZD1839 as
`Second-line therapy
`First-line therapy
`Any prior chemotherapy
`Yes
`No
`Best responses
`CR/PR
`SD
`PD
`CR/PR/SD
`PD
`Skin toxicity
`Yes
`No
`Diarrhea
`Yes
`No
`
`4/10
`21/29
`8/8
`
`22/28
`11/19
`
`18/23
`15/24
`
`27/40
`6/7
`
`4/5
`9/20
`20/22
`13/25
`20/22
`
`11/22
`21/24
`
`16/25
`16/21
`
`13.7
`8.1
`3.2
`
`6.1
`8.1
`
`6.0
`8.6
`
`7.3
`8.4
`
`8.1
`13.7
`3.4
`11.1
`3.4
`
`11.1
`5.3
`
`8.6
`6.1
`
`3.0 to NR
`5.2 to 9.4
`1.0 to 4.1
`
`3.4 to 8.6
`4.3 to NR
`
`3.3 to 8.2
`4.3 to NR
`
`5.0 to 11.1
`1.0 to 9.4
`
`3.3 to NR
`6.7 to NR
`2.9 to 7.3
`8.1 to NR
`2.9 to 7.3
`
`8.3 to NR
`3.4 to 7.3
`
`5.3 to NR
`3.0 to 8.4
`
`90
`62
`13
`
`54
`68
`
`52
`67
`
`58
`71
`
`80
`80
`36
`80
`36
`
`82
`42
`
`68
`52
`
`66
`25
`0
`
`21
`41
`
`21
`39
`
`32
`14
`
`20
`56
`7
`48
`7
`
`49
`13
`
`38
`23
`
`⬍ .0001
`
`.23
`
`.25
`
`.49
`
`.0002
`
`.0001
`
`.001
`
`.12
`
`Abbreviations: CI, confidence interval; CR, complete response; PR, partial response; SD, stable disease; PD, progressive
`disease; NR, not reached.
`*Fraction of patients who have died in the subgroup defined.
`†P value for log-rank test.
`
`epidermal survey. It is possible that the biologic link between
`mucosal surfaces and investigator scoring account for the ob-
`served correlation. The results of a larger ongoing monotherapy
`trial of ZD1839 in SCCHN at the University of Chicago should
`clarify a possible association.
`Not surprisingly, performance status was the strongest predic-
`tor of response and survival. This phenomenon is universal in
`oncology and continues to be the clinician’s best predictor of
`outcome. This study would suggest that patients with poor perfor-
`mance status are unlikely to benefit significantly from this agent.
`A somewhat surprising finding was the observed incidence of
`hypercalcemia. Although this abnormality is well described in
`SCCHN, our experience would suggest that it occurs with a
`lower frequency. It is difficult to attribute this finding to the
`agent per se, although additional experience with ZD1839 should
`address any concerns. Future trials with this agent, especially in
`
`patients with squamous cell carcinomas, should monitor serum
`calcium levels.
`the patient who developed
`In addition to hypercalcemia,
`cervical myelopathy temporally related to administration of
`ZD1839 caused concern. However, we did not observe any other
`neurologic toxicity, and larger trials using this agent have not
`noted similar adverse events. For the most part,
`toxicities
`observed on this trial were commensurate with previous experi-
`ence with this agent.10 Moreover, the lack of any clinically
`obvious differences in toxicity or response end points between
`oral and feeding tube administration suggests that this agent can
`be administered via the latter route without adversely affecting
`patient outcome.
`The laboratory correlative data were somewhat limited be-
`cause of the small number of samples obtained, making conclu-
`sions difficult to draw. The finding of increased ERK staining in
`
`Table 6.
`
`Intensity of Immunohistochemical Staining
`
`Timing*
`
`Pre (n ⫽ 6)
`Post (n ⫽ 4)
`
`0
`
`3
`0
`
`EGFR
`
`2⫹
`
`1
`1
`
`1⫹
`
`0
`0
`
`ERK†
`
`3⫹
`
`2
`3
`
`0
`
`0
`0
`
`1⫹
`
`4
`0
`
`2⫹
`
`3⫹
`
`2
`2
`
`0
`2
`
`0
`
`4
`3
`
`p-ERK
`
`1⫹
`
`1
`0
`
`2⫹
`
`1
`1
`
`3⫹
`
`0
`0
`
`Abbreviations: EGFR, epidermal growth factor receptor; ERK, extracellular signal-related kinase; p-ERK, phosphorylated
`ERK1.
`*Pre samples were obtained before therapy; post samples were obtained at 7 to 8 weeks.
`†Difference between ERK pre- and poststaining was statistically significant (P ⫽ .02).
`
`Downloaded from ascopubs.org by University of Chicago Library, Dr. Mark Ratain on July 27, 2017 from 205.208.061.064
`
`Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
`
`Ex. 1078-0006
`
`

`

`1986
`
`COHEN ET AL
`
`Fig 2. Kaplan-Meier curves of overall survival as a function of (A) performance status (PS), (B) response, (C) disease control, and (D) skin toxicity. CR, complete
`response; PR, partial response; SD, stable disease; PD, progressive disease.
`
`the posttherapy samples is difficult to explain and has not been
`reported by others.32,33 It is possible that this may be related to
`the small number of tumors sampled and random error. One
`would expect a concordant
`increase in p-ERK staining to
`accompany increasing ERK intensity, which was not observed.
`This study was undertaken to demonstrate whether ZD1839
`has activity in SCC