Japanese Journal of Clinical Oncology Advance Access originally published online on September 19, 2005
Japanese Journal of Clinical Oncology 2005 35(10):601-606; doi:10.1093/jjco/hyi151
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2005 Foundation for Promotion of Cancer Research
Primary Tumor of Pancreatic Cancer as a Measurable Target Lesion in Chemotherapy Trials
Division of Hepatobiliary and Pancreatic Medical Oncology, National Cancer Center Hospital East, Kashiwa, Chiba, Japan
For reprints and all correspondence: Hiroshi Ishii, National Cancer Center Hospital East, 6-5-1, Kashiwanoha, Kashiwa, Chiba 277-8577, Japan. E-mail: hirishii{at}east.ncc.go.jp
Received June 16, 2005; accepted August 10, 2005
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Background: It is unclear whether primary pancreatic cancer (PC) tumors can be accepted as measurable target lesions in chemotherapy trials. We reviewed recent PC patients to clarify the significance of their computed tomography (CT) responses of the primary tumor after chemotherapy.
Methods: The patient selection criteria were (i) having been admitted between January 2002 and December 2004, (ii) diagnosed as having histologically or cytologically proven adenocarcinoma of the pancreas, (iii) treated with chemotherapy with no previous anticancer treatment and (iv) having been evaluated by follow-up CT to assess the response according to the Response Evaluation Criteria in Solid Tumors criteria.
Results: A total of 143 patients met the selection criteria. It was possible to measure the largest diameter of the primary tumor in 119 (83%) of the 143, and primary tumor shrinkage was observed in 10 (8%) of the 119. When regarding the primary as measurable as opposed to non-measurable, the number of patients with measurable disease became 127 from 67, and the frequencies of partial response (PR), stable disease (SD) and progressive disease (PD) became 11, 74 and 15% of the 127 from 18, 52 and 30% of the 67, respectively. In the former situation, large primary tumor sometimes canceled the shrinkage or progression of small metastasis. In each setting, PR or SD represented a favorable prognosis compared with PD, however, there were no statistical differences between the PR and the SD.
Conclusion: Measuring the primary tumor is acceptable in 
80% of PC patients. However, we must be aware that the frequency of SD may increase compared with the PR or PD.
Key Words: pancreatic neoplasm • RECIST • computed tomography • measurement • response
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The computed tomography (CT) response to treatment is an important indicator of the therapeutic effect of anticancer agents. In daily practice, response assessment is combined with other indicators of the patient's condition to contribute to the decision-making process. In clinical trials, it is widely used to identify and quantify the antitumor activity of investigational chemotherapy. A valid assessment of the response is based on an accurate measurement of the tumor on CT.
In pancreatic cancer (PC), however, it is difficult to accurately measure the size of the primary tumor mainly due to its invasive growth (1,2). In addition, the appearance of the tumor on a CT scan may not reflect the true proportion of the tumor response due to a vigorous desmoplastic reaction, including inflammation and fibrosis, within and around the tumor (3). Therefore, a primary PC tumor has been regarded as a non-measurable lesion in chemotherapy, and the antitumor effect of CT has been assessed mainly by measurable distant metastasis (4).
Recently, the consensus about the measurability of the primary PC tumor has become unclear. Shrinkage of the primary PC tumor has been more frequently observed since the introduction of gemcitabine (GEM)-based chemotherapy. There have been a number of trials that included not only patients with metastatic disease but also those with locally advanced disease, even though their end point is an objective CT response (5–7). In these reports, the primary PC tumor might be regarded as a measurable lesion, at least in locally advanced PC patients.
The current retrospective study was conducted to clarify whether the primary PC tumor could be accepted as a measurable target lesion in chemotherapy trials.
|
PATIENTS AND METHODS |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The patient selection criteria were (i) having been admitted between January 2002 and December 2004, (ii) diagnosed as having histologically or cytologically proven adenocarcinoma of the pancreas, (iii) treated with chemotherapy with no previous anticancer treatment and (iv) having been evaluated by follow-up CT to assess the response. We listed all patients who met the above criteria and surveyed their records to clarify the significance of the primary tumor measurement in PC chemotherapy.
The tumor response was assessed by CT according to the Response Evaluation Criteria in Solid Tumors (RECIST) response criteria (8). In brief, a complete response was defined as the disappearance of all lesions. A partial response (PR) was defined as at least a 30% reduction in the tumor load, estimated as the sum of the longest diameters of all measurable lesions, taking as a reference the baseline sum of the longest diameters. Progressive disease (PD) was defined as at least a 20% increase in the tumor load, taking as a reference the smallest sum of the longest diameters recorded since the treatment started or development new lesions in a previously uninvolved site. Stable disease (SD) was defined as disease that showed neither sufficient shrinkage nor increase to qualify as either PR or PD.
CT scanning was performed with a four-section multidetector row CT scanner (Aquillion; Toshiba Medical System, Tokyo, Japan). Dynamic contrast-enhanced CT was performed in all patients with the mechanical injection of 100 ml of iopamidol (370 mg/ml of iodine) into the antecubital vein at a rate of 3 ml/s. CT scanning commenced 40–70 s after the start of injection of the contrast medium. Scanning parameters were as follows: 0.5 s gantry rotation time, a beam collimation of 4 x 2 mm, helical pitch of 5 and a reconstruction thickness of 5–7.5 mm.
The CT stage of each patient before chemotherapy was determined prospectively at our film conference held every Tuesday with the attendance of 2–4 staff from each section, i.e. diagnostic radiology, upper abdominal surgery and medical oncology. The measurability of the primary PC tumor on pretreatment CT was judged retrospectively (measurable or non-measurable) by three oncologists (H.I., J.F. and K.N) independently, with no clinical information of the patients. In this retrospective part, the radiologists did not participate in the judgment of measurability for primary lesions. The results were classified into three categories, i.e. ‘measurable’, ‘marginal’ or ‘non-measurable’, the definition of which was as follows: ‘measurable’ when three of the three judged the primary tumor as measurable, ‘marginal’ when two of the three judged the primary tumor as measurable and ‘non-measurable’ when one or none of the three judged the primary as measurable. Finally, the ‘marginal’ cases were determined to be measurable or non-measurable based on the consensus of the three (Fig. 1).
|
In the current study, responses were assessed by two methods, i.e. regarding the primary PC tumor as a non-measurable lesion (referential method) or as a measurable target lesion in cases with ‘measurable’ primary tumors (alternative method). In the latter, the maximum size of the pancreatic mass on CT was measured on the serial axial slices containing the largest portion of the mass. According to the RECIST criteria, shrinkage of the primary PC tumor was defined as 30% or greater reduction of the largest size, and PR was confirmed when the shrinkage continued for more than 4 weeks.
Survival curves were calculated using the Kaplan–Meier method (9). Overall survival was measured from the beginning of chemotherapy to the time of the final follow-up or death. Differences in survival were evaluated with log-rank tests. All analyses were performed using the statistical software SPSS 11.0J for Windows. Statistical significance was defined as a two-sided P-value of 
0.05.
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Between January 2002 and December 2004, a total of 327 histologically confirmed PC patients were admitted to our hospital. Of them, 63 underwent surgical resection, 20 were treated with chemoradiotherapy, 26 received best supportive care and 218 were treated with chemotherapy. Of the 218, 47 had anticancer treatment earlier, i.e. surgical resection of primary PC in 24 and chemotherapy in 23. Of the remaining 171, 28 were not evaluated by CT after chemotherapy because of their early deteriorations. Accordingly, the remaining 143 met the selection criteria in the current study.
The patient backgrounds are shown in Table 1. Of the 143, 67 had measurable metastasis and the remaining 76 had no measurable metastasis. Among them, 101 received GEM monotherapy as a clinical practice (1000 mg/m2/30 min, Day 1, 8 and 15, every 4 week). The remaining 42 received chemotherapy as part of multicenter clinical trials: Phase 1/2 study of the fixed dose rate infusion of GEM (n = 11), and combination of GEM and S-1 (n = 11), Phase 2 study of combination of GEM and infusional fluorouracil (n = 5), S-1 (n = 6), NK911 (micelle forming polymeric doxorubicin, n = 6) and CPT-11 (n = 3).
|
RESPONSE BY THE REFERENTIAL METHOD
Of 67 patients with measurable metastasis, 12 (18%) showed a PR, 35 (52%) remained SD, and 20 (30%) showed PD. All 12 PR cases showed liver metastasis shrinkage. Of 76 with no measurable metastasis, 68 remained SD and 8 showed PD. Therefore, the overall responses in the 143 were 12 PRs, 103 SD and 28 PD, according to a referential method, i.e. regarding the primary PC tumor as a non-target lesion.
RESPONSE BY THE ALTERNATIVE METHOD
The frequencies of patients with a ‘measurable’ primary tumor in the 143 were 76, 85 and 87% according to each of the three blinded reviewers. As a result, the frequencies of ‘measurable’, ‘marginal’ and ‘non-measurable’ primary tumors were 74, 13 and 13%, respectively. Of the 19 ‘marginal’ cases, 13 were reconsidered to be ‘measurable’ cases by discussion. Finally, 119 (83%) were diagnosed as having a measurable primary PC tumor. In the 119, the maximum size of the primary tumor ranged from 21 to 121 mm and the quartiles at 25, 50 and 75% were 37, 47 and 65 mm, respectively. Shrinkage of the primary tumor was observed in 10 of the 119. The relationship between the primary tumor measurability and presence or absence of measurable metastasis is shown in Table 2. Of the 127 with measurable primary or metastatic tumors, 14 (11%) showed a PR, 94 (74%) remained SD and 19 (15%) showed PD. Of the remaining 16 with no measurable lesions, 13 remained SD and 3 showed PD. As a result, the overall responses in the 143 were 14 PRs, 107 SD and 22 PD, according to the alternative method.
|
COMPARISON OF THE TWO METHODS
The relationship between the referential and the alternative responses is shown in Table 3. A discrepancy between the responses was observed in 14 (10%) cases.
|
All five cases from SD (referential) to PR (alternative) had no target metastatic metastasis (locally advanced, 2; minute liver metastasis, 2; peritoneal dissemination, 1), i.e. shrinkage was observed in the primary tumor as the sole target lesion.
Cases from PR to SD, or from PD to SD, had relatively large primary tumors compared with metastasis (Fig. 2). In these cases, the primary tumor was so large to be set off against shrinkage or progression of small metastasis. Therefore, reduction of the total sum of the largest diameter of the targets did not reach over 30% in cases from PR to SD, or the enlargement of the total sum remained within 20% in cases from PD to SD, using the alternative method.
|
Survival curves of measurable cases according to each response are shown in Figs 3 and 4. They showed a situation that was commonly seen in a Phase 2 chemotherapy trial for advanced PC, i.e. a trial for metastatic PC using the referential method and for locally advanced or metastatic PC by the alternative method. PR and SD were superior to PD; however, there were no significant differences between PR and SD in both situations.
|
|
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The newly introduced RECIST, which relies on the single largest dimension of the tumor, is intended to simplify the assessment of the tumor response and has become standard in the world. At the end of the preamble in the RECIST article, there is a statement that specific tumors or anatomic sites presenting unique complexities will be considered in the future. In fact, clinical problems have already been reported in chemotherapy for malignant pleural mesothelioma (10,11) and gastric cancer (12,13) in adopting the unidimensional RECIST criteria. The main cause of the problems arises from the fact that those tumors represent a non-spherical, tridimensional shape at the primary site. PC also demonstrates a non-expanding, invasive growth pattern, and its accurate measuring on CT has been already reported to be difficult (1,2). Therefore, the antitumor effect in Phase 2 chemotherapy trials for PC has been mainly evaluated by the CT assessment of measurable distant metastasis, and primary tumors have generally been regarded as a non-measurable lesion (3).
However, recent reports of Phase 2 trials (5–7) sometimes include not only patients with metastatic disease but also those with locally advanced disease. In patients with locally advanced disease, the primary PC tumor must be measured and assessed using CT. Therefore, the current study focused on the validity of the primary PC tumor measurement and its assessment, because little attention has been given to this issue.
There must be some factors that influence the measurability of the primary PC tumor, such as the quality of CT images, the opinion of each physician and so on. The size of the primary PC tumor may change variously according to the contrast enhanced phase of dynamic CT (1). In the current study, however, an almost uniform method was used to obtain the CT images, thanks to an effort in our diagnostic radiology division. As for the disagreement of each physician's opinion, the current blind test showed a high frequency of agreement about the measurability of the primary tumor. Accordingly, we supposed the primary tumor to be measurable in 80% of advanced PC patients, also in the other center hospitals.
The assessment of primary tumor shrinkage may be used for measuring the anticancer activity in Phase 2 trials for locally advanced PC. In the current study, however, we could not mention this issue because only two patients with locally advanced disease achieved a PR. Overall survival has been usually employed as the primary end point in Phase 2 trials for locally advanced PC (14,15), because standard chemoradiotherapy reproduced almost constant results, i.e. the median survival time of 10 months or 1 year survival rate of 40% (16–18). Accordingly, it may be unnecessary to measure the primary in trials for locally advanced disease.
Measuring the primary PC tumor may be an advantage for recruiting many patients into clinical trials. In fact, patients with measurable lesions increased from 67 to 127 patients. The result indicated candidates for Phase 2 trials may be doubled by measuring the primary tumor. In this manner, however, we should notice that the frequency of SD might increase compared with the PR or PD. As shown in the results, this phenomenon occurred because the large primary tumor sometimes canceled the shrinkage or progression of small metastasis. Accordingly, whether primary tumor can be accepted as a measurable target lesion or not should be determined strictly in each protocol to make an easy interpretation of anticancer activity. To date, the non-PD rate may be the best response indicator, because it is unnecessary to differentiate PR from SD until the development of a new effective chemotherapy superior to standard GEM in the treatment of PC. In this respect, we should be aware of regarding a non-PD rate as an end point, because it may tend to be high using the alternative method in the current study.
In conclusion, the measurement and assessment of the primary PC tumor may be accepted in Phase 2 trials, whereas a careful interpretation of the responses is needed for each protocol.
|
Acknowledgments |
|---|
The authors thank our secretary, Ms Yuki Fujishiro, for her devoted preparation and arrangement of CT imaging.
|
References |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
1 Aoki K, Okada S, Moriyama N, Ishii H, Nose H, Yoshimori M, et al. Accuracy of computed tomography in determining pancreatic cancer tumor size. Jpn J Clin Oncol 1994;24:85–7.
2 Furukawa H, Takayasu K, Mukai K, Inoue K, Kosuge T, Ushio K. Computed tomography of pancreatic adenocarcinoma: comparison of tumor size measured by dynamic computed tomography and histopathologic examination. Pancreas 1996;13:231–5.[Web of Science][Medline]
3 Ahlgren JD. Chemotherapy for pancreatic carcinoma. Cancer 1993; 78:654–63.
4 Okada S. Non-surgical treatments of pancreatic cancer. Int J Clin Oncol 1999;4:257–66.[CrossRef]
5 Philip PA, Zalupski MM, Vaitkevicius VK, Arlauskas P, Chaplen R, Heilbrun LK, et al. Phase II study of gemcitabine and cisplatin in the treatment of patients with advanced pancreatic carcinoma. Cancer 2001;92:569–77.[CrossRef][Web of Science][Medline]
6 El-Rayes BF, Zalupski MM, Shields AF, Vaishampayan U, Heilbrun LK, Jain V, et al. Phase II study of gemcitabine, cisplatin, and infusional fluorouracil in advanced pancreatic cancer. J Clin Oncol 2003;21:2920–5.
7 Arends JJ, Sleeboom HP, Leys MB, Ten Bokkel Huinink D, de Jong RS, Smit JM, et al. A phase II study of raltitrexed and gemcitabine in patients with advanced pancreatic carcinoma. Br J Cancer 2005;92:445–8.[Web of Science][Medline]
8 Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L, et al. New guidelines to evaluate the response to treatment in solid tumors. J Natl Cancer Inst 2000;92:205–16.
9 Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;63:457–81.[CrossRef]
10 Monetti F, Casanova S, Grasso A, Cafferata MA, Ardizzoni A, Neumaier CE. Inadequacy of the new Response Evaluation Criteria in Solid Tumors (RECIST) in patients with malignant pleural mesothelioma: report of four cases. Lung Cancer 2004;43:71–4.[CrossRef][Web of Science][Medline]
11 van Klaveren RJ, Aerts JG, de Bruin H, Giaccone G, Manegold C, van Meerbeeck JP. Inadequacy of the RECIST criteria for response evaluation in patients with malignant pleural mesothelioma. Lung Cancer 2004;43:63–9.[CrossRef][Web of Science][Medline]
12 Koizumi W, Kurihara M, Tanabe S, Kondo I, Yamazaki I, Nonaka M, et al. Advantages of Japanese response criteria for estimating the survival of patients with primary gastric cancer. Gastric Cancer 1999;2:14–9.[CrossRef][Medline]
13 Yoshida S, Miyata Y, Ohtsu A, Boku N, Shirao K, Shimada Y. Significance of and problems in adopting response evaluation criteria in solid tumor RECIST for assessing anticancer effects of advanced gastric cancer. Gastric Cancer 2000;3:128–33.[CrossRef][Medline]
14 Ishii H, Okada S, Tokuuye K, Nose H, Okusaka T, Yoshimori M, et al. Protracted 5-fluorouracil infusion with concurrent radiotherapy as a treatment for locally advanced pancreatic carcinoma. Cancer 1997;79:1516–20.[CrossRef][Web of Science][Medline]
15 Okusaka T, Ito Y, Ueno H, Ikeda M, Takezako Y, Morizane C, et al. Phase II study of radiotherapy combined with gemcitabine for locally advanced pancreatic cancer. Br J Cancer 2004;91:673–7.[Web of Science][Medline]
16 Moertel CG, Childs DS, Reitemeier RJ, Colby MY, Holbrook MA. Combined 5-fluorourail and supervoltage radiation therapy of locally unresectable gastrointestinal cancer. Lancet 1969;2:865–7.[CrossRef][Web of Science][Medline]
17 The Gastrointestinal Tumor Study Group. Therapy of locally unresectable pancreatic carcinoma: a randomized comparison of high dose (6000 rads) radiation alone, moderate dose radiation (4000 rads + 5-fluorouracil), and high dose radiation + 5-fluorouracil. Cancer 1981;48:1705–10.[CrossRef][Web of Science][Medline]
18 Gastrointestinal Tumor Study Group. Treatment of locally unresectable carcinoma of the pancreas: comparison of combined-modality therapy (chemotherapy plus radiotherapy) to chemotherapy alone. J Natl Cancer Inst 1988;80:751–5.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||