Treatment of Neuroendocrine Carcinomas With
`Combined Etoposide and Cisplatin
`Evidence of Major Therapeutic Activity in the
`Anaplastic Variants of These Neoplasms
`Charles G. Moertel, MD, Larry K. Kvols, MD,
`Michael J. O'Connell, MD, and Joseph Rubin, MD
`
`Forty-five patients with metastatic neuroendocrine tumors were treated with a
`regimen of etoposide 130 mg/mz/d for 3 days plus cisplatin 45 mg/m2/d on days 2
`and 3. Both drugs were given by continuous intravenous infusion. Among 27
`patients with well-differentiated carcinoid tumors or islet cell carcinomas, only two
`partial objective tumor regressions were observed (7%). Among 18 patients
`prospectively classified as having anaplastic neuroendocrine carcinomas, however,
`there were nine partial regressions and three complete regressions, an overall
`regression rate of 67%. For anaplastic disease, the median duration of regression
`was 8 months (range to 21 months). Tumor response was unrelated to primary site,
`endocrine hyperfunction, or prior therapy experience. The median survival of all
`patients with anaplastic tumors was 19 months; this seemed favorable when
`considering the small experiences with these rare tumors reported in the literature.
`Toxicity, which was severe for most patients, consisted primarily of vomiting,
`leukopenia, thrombocytopenia, anemia, alopecia, and neuropathy. The anaplastic
`neuroendocrine tumor is strongly responsive to therapy with combined etoposide
`and cisplatin. Patients with undifferentiated carcinomas, originating in typical
`neuroendocrine tumor sites (small and large bowel, pancreas, and stomach) or of
`unknown origin, who have consistent histologic findings by light microscopy
`should be evaluated for this possibility with appropriate immune staining or
`electron microscopy. Cancer 68~227-232,1991.
`
`I special challenge to medical management because they
`
`SLE:T CELL CARCINOMAS and carcinoid tumors are a
`
`are rare, and therefore, sufficient patient numbers are not
`available to conduct the large-scale Phase I1 trials used to
`screen new treatment regimens for activity in the more
`common malignant diseases. Ideas for research trials in
`carcinoid and islet cell carcinoma patients are, therefore,
`often taken from clinical success in more frequently en-
`countered cancers. Certainly the most attractive of these
`would be other cancers derived from neuroendocrine cells,
`i.e., cells with secretory granules and the capability of pro-
`
`From the Department of Oncology, Mayo Clinic, Rochester, Min-
`nesota.
`Supported in part by grant CA 3 I224 from the National Cancer In-
`stitute, National Institutes of Health, Bethesda, Maryland.
`Address for reprints: Charles G. Moertel, MD, Department of On-
`cology, Mayo Clinic, Rochester, MN 55905.
`Accepted for publication April 1, 199 1.
`
`ducing polypeptide hormones or biogenic amines. These
`cells occur in various locations, such as the base of the
`crypts of Lieberkuhn in the appendix and small intestine,
`the islets of Langerhans in the pancreas, and the granular
`basal layer of the bronchial epithelium. The most common
`cancer derived from neuroendocrine cells is the small cell
`carcinoma of the lung.
`Several chemotherapeutic regimens are effective in
`treating small cell lung cancer, not only in producing fre-
`quent objective tumor regressions but also in extending
`patient survival. Probably the most common of these reg-
`imens in modern use are those involving a combination
`of etoposide and cisplatin. This fact led us to the hypoth-
`esis that the activity of etoposide plus cisplatin in small
`cell carcinoma of the lung could predict activity in the
`less common neuroendocrine tumors, i.e., the carcinoid
`and the islet cell carcinoma. In addition, Davis et al.'
`reported activity with this combination in a small group
`of carcinoid tumor patients.
`
`221
`
`NOVARTIS EXHIBIT 2063
`Par v. Novartis, IPR 2016-01479
`Page 1 of 6
`
`

`

`228
`
`CANCER July 15 199 1
`
`Vol. 68
`
`Major differences between the small cell carcinoma and
`the carcinoid or islet cell carcinoma are that small cell
`lung cancer has highly anaplastic histologic findings and
`an aggressive clinical course whereas carcinoids and islet
`cell carcinomas usually are well differentiated and have
`a characteristically indolent course. In the spectrum of
`neuroendocnne neoplasms, there is a group of tumors
`best described as anaplastic neuroendocrine carcinomas.
`These tumors histologically show a less well-developed
`neuroendocrine pattern and have greater cytologic atypia
`than carcinoid tumors or islet cell carcinoma. In addition,
`they have greater mitotic activity and small foci of necrosis
`frequently are present. These anaplastic neuroendocrine
`carcinomas lack the extensive sheet-like necrosis com-
`monly identified in small cell undifferentiated carcinomas.
`In our experience, this histologic finding is associated with
`more rapid advance of malignant disease and less frequent
`clinically recognizable excess hormone production. Be-
`cause the anaplastic neuroendocrine tumors bear a closer
`histologic resemblance to small cell lung cancer than do
`the typical carcinoid or islet cell carcinoma, we also hy-
`pothesized that these anaplastic tumors might be partic-
`ularly vulnerable to etoposide plus cisplatin treatment.
`The particular etoposide and cisplatin schedule we
`chose for study involved a 3 day regimen with each agent
`given by 24-hour intravenous infusion. It was our hope
`that continuous exposure of cancer cells to cisplatin and
`etoposide could enhance the therapeutic interaction of
`these two agents. A previous pilot study of this regimen
`showed that 14 of 15 patients with small cell lung cancer
`had tumor regressions although several of these patients
`were treated with suboptimal doses of etoposide and sev-
`eral had extensive disease.2
`
`Materials and Methods
`Patient Selection
`All patients selected for study had histologic confir-
`mation of metastatic neuroendocrine tumor. Before study
`entry, the patients were classified as having a well-differ-
`entiated carcinoid tumor, a well-differentiated islet cell
`carcinoma, or an anaplastic neuroendocrine tumor. Those
`classified as carcinoids had either an identified primary
`in a typical location (small bowel, appendix, rectum,
`stomach, or lung) or an elevation of urine 5-hydroxyin-
`doleacetic acid (5-HIAA) excretion and a normal pancreas
`on either abdominal exploration, computed tomographic
`scan (CT), or ultrasound examination. Appropriate evi-
`dence for assuming islet cell carcinoma was confirmation
`of a primary tumor in the pancreas by abdominal explo-
`ration, CT scan, or ultrasonography. In the absence of
`clear evidence of such a primary tumor, islet cell carci-
`noma was assumed clinically if the patient had distinct
`elevation of any of the hormonal substances character-
`
`istically produced, e.g., gastrin, glucagon, vasoactive in-
`testinal peptide, or insulin. All patients classified as having
`an anaplastic neuroendocrine tumor were so designated
`by a Mayo Clinic pathologist after review of the histologic
`material. In each instance, this classification was made
`before study entry, not after the fact. This distinction was
`clear, and no tumor was classified as “moderately undif-
`ferentiated.” It was required that each patient have either
`measurable tumor or definite hormonal abnormalities or
`both to serve as indicators of response to therapy. The
`following were accepted as evidence of measurable tumor:
`(1) a clearly demarcated tumor area which was bidimen-
`sionally measurable with a ruler or caliper on physical
`examination or chest radiography, (2) a clearly defined
`perfusion defect on radioisotope liver scan or a tumor
`area on CT scan, either of which must measure at least 5
`cm in greatest diameter, or (3) malignant hepatomegaly
`with a clearly defined liver edge extending at least 5 cm
`below the xiphoid or costal margin on quiet respiration.
`If hormonal abnormalities only were used as indicators
`of response, it was required that pretreatment values be
`at least twice the upper limits of normal. If 24-hour urine
`5-HIAA excretion was used, it was required that the pre-
`treatment value be at least 20 mg (normal, 5 6 mg). It
`was also required that these abnormal values be found on
`two consecutive determinations obtained within 2 weeks
`before study entry. The following were contraindications
`to study entry: leukopenia (< 4000/pl), thrombocytopenia
`(< 1 OO,OOO/pl), an Eastern Cooperative Oncology Group
`(ECOG) performance score of 3 or 4 (unless disability was
`caused solely by a hormonal syndrome), uncontrolled in-
`fection, immunotherapy, or chemotherapy during the
`preceding 3 weeks, any prior therapy with either etoposide
`or cisplatin, and a serum creatinine greater than 1.5
`mgldl.
`
`Treatment Method
`
`Before therapy, and also at the time of each evaluation,
`a medical history was taken, and the patient underwent
`physical examination and tumor measurements. Labo-
`ratory analysis included leukocyte count, platelet count,
`hemoglobin, a blood chemistry panel, and assays of hor-
`monal indicators. Appropriate imaging was obtained of
`any indicator lesions, as was chest radiography, pretreat-
`ment and at every other posttreatment evaluation.
`Therapy was administered in the hospital. Etoposide
`was administered by continuous infusion in a solution of
`5% dextrose and 0.45% saline. The dose was 130 mg/m2
`given daily for 3 consecutive days. Cisplatin was given in
`the same infusion at a dose of 45 mg/m2 on days 2 and
`3 of etoposide administration. Appropriate antiemetics
`were administered with each course of therapy. Courses
`were repeated every 4 weeks with reductions in dose if
`
`NOVARTIS EXHIBIT 2063
`Par v. Novartis, IPR 2016-01479
`Page 2 of 6
`
`

`

`No. 2
`
`ETOPOSIDE AND CISPLATIN FOR NEUROENDOCRINE CA
`
`* Moerfel et a/.
`
`229
`
`excessive toxicity was experienced during the preceding
`course. Therapy was continued until tumor progression,
`or in the case of stable disease or tumor regression, as
`long as there was no evidence of symptomatic or general
`deterioration, and the therapy was tolerable clinically.
`All patients were evaluated for therapeutic response
`before initiation of their next course of therapy. Patients
`were declared to have had a complete regression if there
`was total disappearance of all clinically detectable tumor
`and return of all abnormal hormonal indicators to normal
`range. Partial regression was defined as a greater than 50%
`reduction in the products of the longest perpendicular
`diameters of lesions measured by either physical exami-
`nation or imaging. If malignant hepatomegaly was the
`indicator, it was required that there be more than a 30%
`decrease in the sum of distances below the costal margin.
`To declare a hormonal response, it was required that this
`parameter be reduced to less than 50% of the pretreatment
`value or to normal range. Time to progression and survival
`
`were measured from the time of study entry. Duration of
`tumor regression was measured from time of entry to the
`last time regression was documented.
`
`Results
`A total of 46 patients, all eligible according to the study
`criteria, were enrolled between May 1987 and February
`1990. A single patient withdrew consent and left the study
`after only 1 day of therapy. We were unable to obtain any
`follow-up information on this patient. He therefore was
`considered not evaluable, and he is not included in any
`of the following analyses. Among the 45 eligible and eval-
`uable patients, 13 had well-differentiated carcinoid tu-
`mors, 14 had well-differentiated islet cell carcinomas, and
`18 had anaplastic neuroendocrine tumors. Anaplastic tu-
`mors were identified as neuroendocrine by consistent his-
`topathologic findings plus a clinical endocrine syndrome
`in ten patients, by light microscopy only in one patient,
`
`TABLE I. Patient Characteristics
`
`Carcinoid
`tumor
`
`13
`815
`61 (52-67)
`
`28
`(%-loo)
`-
`5
`2
`3
`I
`-
`2
`
`12
`-
`-
`-
`-
`-
`-
`1
`9
`
`12
`-
`12
`
`Islet cell
`carcinoma
`
`14
`717
`47 (22-64)
`
`20
`(1-41)
`
`14
`-
`-
`-
`-
`-
`-
`
`-
`6
`5
`3
`2
`2
`-
`4
`10
`
`1 1
`3
`9
`
`Anaplastic
`neuroendocrine
`carcinoma
`
`18
`10/8
`52 (24-74)
`
`4.5
`(5 days-41 mo)
`
`6
`2
`1
`1
`2
`3
`3
`
`5
`2
`I
`-
`1
`-
`2
`9
`3
`
`14
`7
`7
`
`Characteristic
`
`Total no. of patients
`Malejfemale
`Median age in yr (range)
`Median time from diagnosis of metastasis
`in mo
`(Range)
`Primary tumor
`Pancreas
`Small bowel
`Lung
`Cecum & right colon
`Rectum
`Stomach
`Unknown
`Endocrine abnormalities*
`5-HIAA
`Gastrin
`Glucagon
`Insulin
`Pancreatic polypeptide
`Hypercalcemia
`ACTH
`None
`Prior chemotherapy
`Indicators of response*
`Liver tumor
`Other tumor
`Endocrine marker
`Performance score?
`0
`1
`2
`3
`
`--
`5-HIAA: 5-hydroxyindoleacetic acid.
`* Several patients had more than one endocrine abnormality or in-
`dicator of response.
`
`4
`8
`1
`-
`
`6
`5
`5
`7
`3
`5
`-
`I $
`t ECOG performance score: 0 (fully active) to 4 (totally disabled).
`$ Disability caused by endocrine syndrome.
`
`NOVARTIS EXHIBIT 2063
`Par v. Novartis, IPR 2016-01479
`Page 3 of 6
`
`

`

`230
`
`CANCER July 15 199 1
`
`Vol. 68
`
`and in seven patients, by light microscopy plus electron
`microscopy (five patients) and/or immune staining (four
`patients). In the latter group, two tumors were positive
`for neuron-specific enolase, two for synaptophysin, and
`two for chromogranin.
`The characteristics of these patients are documented
`in Table 1. Patients with islet cell carcinomas and ana-
`plastic neuroendocrine tumors were younger than patients
`with carcinoids. Most had primary tumors in areas typical
`for neuroendocrine neoplasms. The five patients classified
`as having tumors of unknown origin had been examined
`with chest radiography, CT scanning, gastrointestinal im-
`aging procedures, and in one case each, laparoscopy and
`surgical exploration of the abdomen. Patients with ana-
`plastic neuroendocrine tumors had a much shorter inter-
`val from diagnosis of metastatic disease to the onset of
`therapy, one half of them did not have associated endo-
`crine syndromes, and most had had no prior chemother-
`apy exposure. Seventy-eight percent of patients entered
`in this trial had an excellent or good performance score
`(ECOG score, 0 or 1).
`
`Therapeutic Results
`
`Patients with typical well-differentiated carcinoid tu-
`mors and islet cell carcinomas were treated with a median
`of three courses of therapy (range, one to seven courses),
`whereas patients with anaplastic neuroendocrine tumors
`were treated with a median of five courses (range, two to
`22 courses).
`Response to therapy according to the three tumor types
`is displayed in Table 2. Among the 13 patients with well-
`differentiated carcinoid tumors, none showed an objective
`tumor response, reduction in 5-HIAA level, or improve-
`ment of their carcinoid syndrome. Eleven patients had
`stable disease, but the median duration of this stability
`
`was only 3 months, and the median interval to progression
`for all patients was only 3 months.
`Among the 14 patients with islet cell carcinoma, two
`(1 5%) had partial tumor regressions. Both of these patients
`had had prior chemotherapy. One of these with a non-
`functioning tumor showed a greater than 50% reduction
`in abdominal nodal masses displayed by CT scan; this
`persisted for 4.5 months. The second patient had a greater
`than 50% reduction in abdominal and liver masses and
`reduction in gastrin level from 480 to 110 pg. This re-
`sponse persisted for 6 months. It is interesting that one
`patient with tumor progression and increasing levels of
`gastrin and glucagon also had complete relief of hyper-
`insulinism and accompanying intractable hypoglycemia.
`As with the carcinoid tumor, both duration of stable dis-
`ease and interval to progression were short: 3 months and
`4 months, respectively.
`In contrast, 12 of 18 patients (67%) with anaplastic
`neuroendocrine pathologic findings had objective tumor
`regression. Three of these tumors regressed completely.
`In two this was determined by clinical findings and im-
`aging and in one by restaging laparotomy with biopsies
`of previously involved organs. Regression was evidenced
`quickly (median, 1 month). The overall median duration
`of regression was 8 months (range to 21 months). In all
`instances regression involved reduction in measurable tu-
`mor mass. Response was not perceptibly influenced by
`tumor function. Among seven patients with endocrine
`response indicators, four had regression. Two of these pa-
`tients had reduction of elevated 5-HIAA levels to normal
`range, and one had a 5-HIAA reduction from 3 1 1 mg (in
`24 hours) to 103 mg. In one patient elevations of both
`gastrin and glucagon returned to normal. Two of the three
`patients with prior chemotherapy exposure had partial
`responses. The median interval to progression for all an-
`aplastic tumor patients was 9 months (range to 21
`months).
`
`TABLE 2. Objective Response to Therapy
`
`Carcinoid tumor
`(n = 13)
`- -
`1 1
`2
`
`-
`-
`
`3
`1-21
`
`10%
`3-36+
`
`Islet cell carcinoma
`(n = 14)
`-
`2 (14%)
`9
`3
`
`5
`4l/2 and 6
`
`4
`1-8
`
`15%
`4-36%+
`
`Anaplastic neuroendocrine
`carcinoma
`(n = 18)
`
`3 (17%)
`9 (50%)
`6
`-
`
`8
`3-2 I
`
`1 1
`2-2 1
`
`19
`5-36+
`
`Complete regression
`Partial regression
`Stable
`Progression
`Duration of regression*
`Median (mo)
`Range (mo)
`Interval to progression*
`Median (mo)
`Range (mo)
`SUNiVd*
`Median (mo)
`Range (mo)
`* Estimates: Kaplan-Meier method.
`
`NOVARTIS EXHIBIT 2063
`Par v. Novartis, IPR 2016-01479
`Page 4 of 6
`
`

`

`No. 2
`
`ETOPOSIDE AND CISPLATIN FOR NEUROENDOCRINE CA
`
`* Moertel et Ul.
`
`23 1
`
`TABLE 3. Toxic Reactions
`
`Survival
`Twenty-eight of our 45 patients died at a median time
`of 12.5 months after study entry (range, 3 to 28 months).
`Seventeen are still living at a median time of 14.5 months
`(range, 1 1 to 36.5 months). In Figure 1 we plotted patient
`survival by the Kaplan-Meier method according to tumor
`classification and measured from the time of study entry.
`The median survival for the well-differentiated neuroen-
`docrine tumors was 15 months (carcinoid tumor, 10.5
`months and islet cell carcinoma, 15.5 months). Anaplastic
`neuroendocrine carcinomas showed a modest advantage
`with a median of 19 months.
`
`Toxicity
`Drug toxicity was a major problem for most patients.
`This frequently required dosage reductions or cessation
`of therapy. The toxic reactions are documented in Table
`3. Nausea and vomiting were frequent despite vigorous
`antiemetic therapy with metoclopramide, dexamethasone,
`and other agents. These reactions, however, usually were
`not severe and did not interfere significantly with patient
`compliance. Mucocutaneous reactions and diarrhea were
`relatively minor problems. All patients noted alopecia,
`usually near complete. Neuropathy, usually sensory and
`presumably related to cisplatin, was experienced by 24%
`of patients; in 7%, it was severe. Hearing loss was infre-
`quent. Two thirds of patients had renal toxicity as evi-
`denced by a rising creatinine level. This rise was usually
`mild (< 2 mg/dl), but one patient had a complete renal
`shutdown, requiring dialysis for several weeks. A single
`patient had severe chills and fever with negative blood
`cultures; we presumed this was a reaction to etoposide.
`Therapy was not continued in this patient.
`Hematologic toxicity was universal. All patients had
`leukopenia. In five patients, this was life threatening
`
`100
`
`80
`
`-
`'5
`$ 60
`-
`k 40
`
`m
`
`20
`
`0
`
`- anapiaslic (N-18)
`
`well differentialed (N-27)
`
`1
`
`!
`
`L
`
`0
`
`6
`
`12
`
`24
`
`30
`
`18
`months from study entv
`FIG. 1. Comparative survivorship: well-differentiated and anaplastic
`neuroendocrine tumors.
`
`Nonhematologic toxicity
`Toxic reaction
`Nausea
`Vomiting
`Diarrhea
`Stomatitis
`Neuropathy
`Dermatitis
`Chills & fever
`Hearing loss
`Alopecia
`Renal (f creatinine)
`Hematologic toxicity
`
`Percent of 45 patients
`
`Any
`96
`93
`16
`13
`24
`4
`2
`L
`100
`66
`
`Severe*
`13
`9
`-
`2
`I
`-
`2
`-
`I00
`2
`
`Percent of 44 patientst
`
`35
`53
`12
`
`49
`14
`21
`
`Toxic reaction
`Leukopenia (cells/pl)
`< 4000 r 2000
`< 2000 t 1000
`< 1000
`Thrombocytopenia (cells/pl)
`< 130,000 2 50,000
`< 50,000 t 25,000
`< 25,000
`Anemia (fall in g of Hgb)
`38
`r 1 < 3
`29
`r 3 < 4
`2 4
`22
`* Grade 3 or 4 according to NCI Common Toxicity Critena.
`t Adequate counts not obtained in one patient.
`
`~
`
`~~
`
`~
`
`~
`
`~
`
`~
`
`(< lOOO/pl), and in three patients, it was complicated by
`sepsis. Thrombocytopenia was usually mild, but nine pa-
`tients had counts less than 25,000 pl. No patient hem-
`orrhaged, but one had cutaneous purpura. Thirteen pa-
`tients had worsening leukopenia and/or thrombocyto-
`penia after repeated courses of therapy. Anemia occurred
`consistently with repeated courses of therapy and fre-
`quently was symptomatic and associated with easy fatigue
`or exertional dyspnea. There was a definite tendency to
`increasing anemia with increasing duration of therapy.
`Overall, most patients had either severe or life-threatening
`toxicity, but there were no drug-related deaths.
`
`Discussion
`
`Our results identify a neoplasm that is highly responsive
`to chemotherapy. The overall regression rate of 67% with
`a complete regression rate of 17% found with etoposide
`and cisplatin therapy of anaplastic neuroendocrine car-
`cinomas was comparable to that in extensive small cell
`lung cancer. The duration of response was substantive.
`Although impossible to prove in the absence of a ran-
`domized trial, therapy may add to life expectancy in these
`patients. Their projected median survival was 19 months
`from the onset of treatment. There is little survival infor-
`mation regarding these patients in the literature, but a
`median survival of 6 months was reported in a small group
`
`NOVARTIS EXHIBIT 2063
`Par v. Novartis, IPR 2016-01479
`Page 5 of 6
`
`

`

`232
`
`CANCER July 15 199 1
`
`Vol. 68
`
`of patients with carcinoid tumors with an undifferentiated
`growth at tern.^ A similar median survival of 7 months
`was documented in patients with moderate to poorly dif-
`ferentiated neuroendocrine carcinomas of the colon and
`r e ~ t u m . ~
`The responsiveness to etoposide and cisplatin therapy
`was not seen in our 27 patients with well-differentiated
`neuroendocrine tumors who had an overall objective
`regression rate of only 7%. There were other differences
`between these patient populations, of course, most prom-
`inently a much larger proportion of patients with well-
`differentiated tumors who had had prior chemotherapy.
`However, no regressions were observed in eight well-dif-
`ferentiated tumors in patients who had had no prior che-
`motherapy exposure. With regard to other features that
`commonly influence response to chemotherapy (such as
`specific organ involvement, extent of disease, and perfor-
`mance status), the two patient groups were comparable.
`Another study was pertinent to our trial.5 Among a
`population of patients with poorly differentiated carci-
`noma of unknown origin, these authors found 29 patients
`whose tumors had neuroendocrine features. In contrast
`to our observations, their patients had largely retroperi-
`toneal or peripheral nodal involvement with only four
`showing hepatic metastasis. One of their patients was
`found at necropsy to have small cell lung cancer, and they
`speculated this may have been true for several others.
`None of their patients had clinical evidence of excessive
`hormonal production, but one patient had a carcinoid
`tumor at necropsy. Although these contrasts may indicate
`fundamentally different diseases, it would seem more
`likely that both studies describe different shades of the
`anaplastic neuroendocrine tumor spectrum with ours
`closer to the typical neuroendocrine carcinoma and theirs
`closer to small cell lung cancer. In their therapeutic ex-
`perience, five of six patients responded to etoposide and
`cisplatin, and 12 of 13 patients responded to other cis-
`platin-containing regimens.
`Immunohistochemical and other techniques for rec-
`ognition of neuroendocrine features are now available. In
`the past, these tumors may have been considered undif-
`ferentiated adenocarcinomas. With the recognition of ef-
`fective therapy for these neoplasms, it would now seem
`reasonable to consider this possibility in clinical circum-
`stances that would otherwise be therapeutically intracta-
`ble, e.g., highly undifferentiated carcinoma of the pan-
`creas. It is possible that the anaplastic neuroendocrine
`carcinoma may be more common than believed in the
`
`past. When 50 consecutive colorectal carcinomas origi-
`nally classified as “poorly differentiated” were studied,
`the authors found that approximately 50% of them had
`the immunostaining characteristics of neuroendocrine
`~arcinorna.~ Among 220 patients with highly anaplastic
`carcinomas of unknown origin, others found 13% to have
`neuroendocrine feature^.^ It seems appropriate for the pa-
`thologist to pursue the possibility of anaplastic neuroen-
`docrine carcinoma in patients (1) who have a compatible
`histologic examination by light microscopy, (2) who have
`primary tumors in organs that most commonly give rise
`to neuroendocrine carcinomas, e.g., stomach, pancreas
`and small or large bowel, or (3) who have undifferentiated
`carcinomas of unknown origin. For this purpose, helpful
`techniques include electron microscopy or staining for
`the so-called panneuroendocrine markers, i. e., synapto-
`physin, chromogranin or neuron-specific enolase.
`The specific etoposide and cisplatin regimen we used
`was cumbersome and expensive because of the constant
`infusion method requiring hospitalization. There is only
`tenuous theoretic evidence that this method may be more
`effective than the much less costly rapid injection regimen
`for this drug combination that has been in common use
`and can be easily handled on an outpatient basis. Others
`obtained comparable results with rapid injection, and this
`would seem to be the preferred m e t h ~ d . ~
`Toxicity was a major problem in our trial with frequent
`severe leukopenia and chronic reactions of anemia and
`neuropathy. Whereas it is possible that a less intense reg-
`imen would be equally effective, this assumption is un-
`proven. Nevertheless, consideration might be given to
`more moderate doses, alternating courses of the two-drug
`combination with courses of etoposide alone to reduce
`total cisplatin exposure, and discontinuing therapy after
`a maximum response has been obtained.
`
`REFERENCES
`1. Davis S, Tonato M, Crini L ef al. Treatment of metastatic carcinoid
`and Merkel cell tumors with cisplatin and etoposide (Abstr). Proc Am
`SOC Clin Oncol 1987; 693.
`2. b o o k JE, Jett JR, Little C. A phase 1-11 study of sequential infusion
`VP- 16 and cisplatin therapy in advanced lung cancer. Am J Clin Oncol
`1989; 12:114-117.
`3. Johnson LA, Lavin P, Moertel CG ef al. Carcinoids: The association
`of histologic growth pattern and survival. Cancer 1983; 5 1:882-889.
`4. Staren ED, Could VE, Warren WH ef al. Neuroendocrine carci-
`nomas of the colon and rectum: A clinicopathologic evaluation. Surgery
`1988; 104:1080-1089.
`5. Hainsworth JD, Johnson DH, Greco FA. Poorly differentiated
`neuroendocrine carcinoma of unknown primary site. Ann Intern Med
`1989; 109:364-371.
`
`NOVARTIS EXHIBIT 2063
`Par v. Novartis, IPR 2016-01479
`Page 6 of 6
`
`