NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Cancer (UK). Diagnosis and Management of Metastatic Malignant Disease of Unknown Primary Origin. Cardiff (UK): National Collaborating Centre for Cancer (UK); 2010 Jul. (NICE Clinical Guidelines, No. 104.)

Cover of Diagnosis and Management of Metastatic Malignant Disease of Unknown Primary Origin

Diagnosis and Management of Metastatic Malignant Disease of Unknown Primary Origin.

Show details

Appendix BEconomic Plan

NATIONAL INSTITUTE FOR HEALTH AND CLINICAL EXCELLENCE

This document identifies the priorities for economic analysis and the proposed methods for addressing these questions as described in section 8.1.3.1 of the Guidelines Manual (2006)

1. Guideline

Title of guideline: Metastatic malignant disease of unknown primary origin

2. Process for agreement

The economic plan was prepared by the guideline economist in consultation with the rest of the NCC technical team and GDG. It was discussed and a agreed on 06-02-2009 by the following peoplea.

For the NCC and GDG

NCC economist:Eugenia Priedane
NCC representative(s)bAndrew Champion
Victoria Titshall
Angela Bennett
GDG representative(s)c:Andy Fowell (GDG chair)
Richard Osborne (GDG Lead Clinician)

For NICE

CCP lead d:
Commissioning manager:Nicole Elliott
Economic leade:Francis Ruiz/ Stefanie Kinsley
Costing lead:

Proposals for any substantive changes will be circulated by email to this group. If revisions are agreed, they will be listed as addenda to this document (section 5 below),

Proposed economic plan

The purpose of this document is to highlight our deliberations in terms of setting the priorities for undertaking economic analyses to inform guideline development for metastatic malignant disease of unknown primary origin, or cancer of unknown primary (CUP). Previous conversations at GDG meetings have highlighted the challenges of conducting economic evaluations in this disease area, mostly because of the heterogeneity of the patient population and a lack of clinical effectiveness data in relation to interventions and procedures for CUP patients. More specifically it is considered unlikely, given the general paucity of evidence on CUP-related technologies, that we will be able to produce estimates of cost-effectiveness that are robust and reliable.

After discussions with NICE (and the Clinical Lead and Chair), we propose that development of de novo economic models may be used to estimate the value of future research for CUP-related technologies for which there may be a high level of uncertainty about the incremental effectiveness and/or cost between two or more clinical strategies. This type of analysis could help inform research recommendations within the guideline. The table below summarises and identifies topics that have been prioritised for this approach.

Clinical TopicRequires analysis?Comments and explanation
Topic 1

For patients with malignancy of undefined primary origin, is the existing screen of basic investigations effective in identifying the maximum number of possible primary cancers as rapidly as possible?
Not relevantThe shortage of studies comparing different strategies, or examining the individual contribution of individual tests, compromises the value of this PICO.
An agreement has been reached among the GDG that some form of consensus-based recommendation be included in the guideline. Given these limitations, this topic is not suitable for economic evaluation.
Topic 3

For patients with malignancy of undefined primary origin undergoing initial diagnostic tests, is there benefit in terms of patient outcomes or speed through the process of doing serum tumour markers?
LowThis topic potentially affects all CUP patients. Tumour markers are a relatively inexpensive diagnostic tool. There is no clear evidence on the sensitivity and specificity of tumour markers and their subsequent effect on health outcomes. Moreover, Varadhachary et al. states that tumour markers play more of a prognostic rather than a diagnostic role. (Varadhachary et al. 2004). Therefore, it is not clear what (if any) health benefits are associated with tumour markers. Given these limitations, it would be difficult to conduct a meaningful economic evaluation.
Topic 4
For patients with malignancy of undefined primary origin, is the use of upper- and lower-GI endoscopy in all patients more effective in identifying the maximum number of possible primary cancers as rapidly as possible compared with selective, symptomdirected GI endoscopy?
LowThere are no economic studies and there is limited clinical evidence on the use of upper- and lower-GI endoscopy in patients with a provisional diagnosis of malignancy of unknown primary who are undergoing initial diagnostic tests and its effectiveness in identifying the primary site. Given these limitations, it would be difficult to conduct a meaningful economic evaluation.
Topic 5
For patients with malignancy of undefined primary origin does evaluation by a specialist oncology team at an earlier time than is traditionally the case improve outcomes?

Topic 6
Is consistent support from an identified key worker, e.g. a specialist nurse, from the point a patient is diagnosed with an unknown or uncertain primary cancer, more effective than no support?

Topic 7
For patients with malignancy of undefined primary origin, is it beneficial for investigations to be undertaken to end uncertainty when there is little likelihood of clinical benefit?

Topic 10
For patients with malignancy of undefined primary origin undergoing screening investigations to identify a primary site, does management by a specialist CUP MDT result in greater benefits than the existing non- MDT management?
LowDefining benefit or measuring effectiveness in quantitative terms for these topics is challenging. For each topic, it will be difficult identify discrete pathways and consequences/outcomes of each strategy or service configuration in a manner that would lend itself to economic evaluation or decision analytic techniques. The availability of economic evidence to inform these topics is also low. Given these limitations, this topic is not suitable for economic evaluation.
Topic 8

For patients with malignancy of undefined primary origin, does immuno-histochemical analysis (using CK20 and CK7) to define putative tissue/organ of origin, or hormone receptor (ER, PR) analysis to potentially predict response to hormonal therapies, result in improved outcomes?
LowBased on previous discussions with RO and AF, the aim of this topic is to highlight a gold standard of managing patient in each of these subgroups, i.e. the optimal use of immuno-histochemical analysis.
No evidence has been found to indicate a change in overall patient outcomes. Lack of evidence paired with relatively inexpensive competing alternatives does not render this topic a high priority.
Topic 11

For patients with provisional Cancer of Unknown Primary with clinical features compatible with metastatic breast cancer, does contrast-enhanced breast MRI improve detection of occult primary breast cancer?
LowThe proportion of patients covered by this topic is limited (around 500 patients). Furthermore there are no economic studies and there is limited clinical evidence on the effect of the contrast-enhanced breast MRI in patients with unknown primary on their overall survival.
Given these limitations, it would be difficult to conduct a meaningful economic evaluation.
Topic 13
For patients with provisional Cancer of Unknown Primary does PET-CT result in improved outcomes?
HighIf evaluated at the beginning of the patient pathway, this topic could potentially affect all patients with cancer of unknown primary. FDG PET-CT has shown to have a higher sensitivity and specificity, particularly for patients presenting with cervical lymphadenopathy (Fogarty et. al. 2003, Johansen et. al. 2008). There are cost implications of implementing wider use of PET-CT, but identification of the primary tumour using this imaging technique can also lead to a change in treatment decision that may result in improved patient outcomes. There are no prospective studies that have evaluated subsequent impact of PET-CT on patient outcomes, however an economic analysis using the expected value of information approach can be used to quantify the level of uncertainty and inform research recommendations into the value of PET-CT in CUP patients.
Topic 14

For patients with confirmed Cancer of Unknown Primary who present with brain metastases, does specific treatment guided by putative site of primary origin improve outcomes, compared with generic treatment comprising supportive care ± palliative radiotherapy?
LowCUP patients with brain metastases have a particularly poor prognosis and chemotherapy has limited efficacy in brain metastases. Cranial irradiation may be used to provide symptomatic care. There is little available data to quantify the benefit of cranial irradiation beyond symptomatic relief. In the absence of more effective active treatments, and given that this topic affects a small proportion of CUP patients, no economic evaluation will be undertaken.
Topic 17

For patients with provisional Cancer of Unknown Primary who present with intra-pulmonary nodules without evidence of endobronchial disease, does bronchoscopy result in improved outcomes?
LowThis topic affects about 10% of CUP patients. There are no economic studies and there is limited clinical evidence on the effect of performing a bronchoscopy in this patient group on overall survival, thereby limiting the feasibility of conducting an economic analysis.
Topic 19

For patients with provisional Cancer of Unknown Primary who present with ascites, does cytological examination of ascitic fluid, or histological examination of malignant peritoneal tissue result in a superior clinical outcome?
LowThere are no economic studies and there is limited clinical evidence on the effect of the cytological examination of ascitic fluid in this patient group on overall survival. Given these limitations, this topic is not suitable for an economic evaluation.
Topic 21
What is the optimal management for patients with confirmed Cancer of Unknown Primary who present with squamous carcinoma involving upper / mid neck nodes?

Topic 22
What is the optimal management for patients with confirmed Cancer of Unknown Primary who present with adenocarcinoma involving axillary nodes?

Topic 23

What is the optimal management for patients with confirmed Cancer of Unknown Primary who present with squamous carcinoma involving inguinal nodes?
LowTopics 21, 22 and 23 affect a relatively small proportion of patient in each subgroups of patients (about 5%).
This topic is considered low priority since the number of patients affected is not as great as for other topics. Furthermore, the level of clinical evidence would be crucial in conducting a meaningful economic evaluation.
Topic 24
What is the benefit of radical local treatment for patients with confirmed Cancer of Unknown Primary who present with an isolated metastasis in one of the following organs: brain, bone, liver, skin, lung?

Topic 25
For patients with confirmed Cancer of Unknown Primary in whom systemic treatment is being considered, are there prognostic factors that significantly influence outcome and which should be considered in treatment decisions?
LowThe patient subgroups covered in this topic are very small. Based on previous discussions with RO and AF, the aim of this topic is to highlight a gold standard of managing patients in each of these subgroups.

No evidence has been found to indicate a change in overall patient outcomes. Lack of evidence paired with very small subgroups of patients render this topic a low priority.
Topic 26

For patients with confirmed Cancer of Unknown Primary in whom systemic treatment is being considered, does gene expression-based classification (according to putative tissue of origin) lead to improved outcomes (through the use of treatment chosen on the basis of the predicted primary site)?
LowIn recent years gene expression profiling has demonstrated the ability to identify a broad spectrum of tumour types at the molecular level. Furthermore, CUP gene expression analysis can be utilized in conjunction with, or in place of standard investigative diagnostic procedures to expedite the diagnostic process of a cancer of unknown primary. (Buckhaults et. al. 2003, Bridgewater et. al. 2008)

Although this topic could all CUP patients, at present in the UK, gene expression-based classification is not common practice and there is no data to estimate resource use and health benefits associated with this diagnostic option.

Given these limitations, it would be difficult to generate a meaningful estimate of cost-effectiveness and is considered a low economic priority.
Topic 27

For patients with confirmed Cancer of Unknown Primary with no clinical features fitting a recognised syndrome, in whom systemic treatment is being considered, does treatment improve the outcome, compared with symptomatic treatment alone?
HighThere is little high quality evidence on the benefit of chemotherapy in CUP patients with no clinical features fitting a recognised syndrome. However, given the additional costs of active treatment, an economic analysis may help reduce the level of uncertainty around the use of chemotherapy over and above supportive care in these patients. This topic was highlighted as an economic priority at the 4th GDG meeting.
Topic 28
For patients with confirmed Cancer of Unknown Primary in whom systemic treatment is being considered, if clinical features match a recognised syndrome, does treatment guided by that syndrome result in better outcomes than generic treatment?
MediumFor patients with confirmed CUP with clinical features matching a recognised syndrome, the use of chemotherapy is more established than for the patient population in Topic 27. Topic 28 relates to the decision between various chemotherapy regimens for 5 specific subgroups of CUP patients, whereas Topic 27 addresses the value of chemotherapy over and above best supportive care in patients with no clinical features matching a recognised syndrome. Topic 27 may inform the comparator for Topic 28, therefore it should be considered first for economic analysis. Furthermore, Topic 27 is considered a higher priority for economic analysis because there is a higher need to address the uncertainty surrounding the cost-effectiveness of chemotherapy versus best supportive care. To meet the time constraints of the guideline development process, Topic 27 has been prioritised over Topic 28.

For each question where economic analysis is proposed:

Question number(s)Outline proposed method of analysis
Topic 13Proposed analysis

Aim
An analysis will be carried out to assess the value of perfect information of carrying out PET-CT scan for patients with provisional cancer of unknown primary and negative initial work up, which included whole body CT and biopsy where appropriate. The findings of this analysis will be used to inform future research recommendations.

Intervention

PET-CT scan performed after negative initial diagnostic work up (including whole body CT scan and biopsy where appropriate.

Comparators

No PET-CT

Methods

PET-CT has shown to have a higher sensitivity and specificity, particularly for patients presenting with cervical lymphadenopathy (Fogarty et. al. 2003, Johansen et. al. 2008).
Decision analysis will be used to model the clinical pathway and to estimate the expected value of perfect information of performing PET-CT in patient with provisional CUP patient who have negative initial diagnostic work up compared to not doing PET-CT will be considered in this analysis.
Formal value of information analysis will provide an analytic framework which will address whether a decision on whether to adopt PET-CT scanning after initial negative diagnosis can be made on the basis of current evidence or whether more evidence is required to support the decision about PET-CT scanning in the future, and how much we should be prepared to pay for this evidence. (Claxton et al. 2002)

There are three components in this framework:
“(i) the construction of a decision analytic model to represent the decision problem; (ii) a probabilistic analysis of this model to characterize the current decision uncertainty; and (iii) establishing the value of additional information” (Briggs et al. 2006)

The structure of the analytic model will be informed by the data available in the literature and in consultation with the GDG.

The NHS perspective will be adopted; that is the health benefits and costs to be considered in the analysis will only be those relevant to the NHS. Relevant costs include those borne by Personal Social Services (PSS) as well as those that fall on the NHS itself. Unit costs will be derived from publicly available national sources whenever possible (e.g. NHS Reference Costs).

Probability distributions will be assigned to different clinical and cost parameters within the model so that a probabilistic sensitivity analysis can be carried out to assess the overall uncertainty of the model and the robustness of the results.

Feasibility issues

Given the heterogeneity of the CUP patient population, it is not clear if a single model will be able to accommodate (mathematically) all relevant testing options and possible test results to accurately reflect clinical reality. This will require close discussions with the clinical contact and GDG to reach agreement on the appropriateness and feasibility of conducting the analysis.

It is unlikely that the literature will be sufficient to populate all relevant parameters in the model. Expert elicitation will be considered where estimates from the literature are not available.

Update following 8th GDG
  • The team of health economists made several attempts to draft an economic model structure for Topic 13 that is both clinically accurate and methodologically feasible for undertaking a decision analytic approach to which EVPI can be applied.
  • Each version of the model was discussed in conjunction with the designated clinical contact from the GDG.
  • Initial clinical guidance suggested that due to the heterogeneity of the CUP population, separate models should be considered for subgroups of CUP patients because
    • The possible outcomes of PET-CT will differ depending on the patient’s initial distribution of metastases. It was considered desirable to factor in >5 possible consequences (i.e. in the context of CUP, PET-CT is viewed as a diagnostic test that does not have a binary outcome). Each outcome could result in set of follow-up tests and potentially different treatment decisions leading to different survival and QALY estimates.
    • For the comparator (do not conduct PET-CT arm in the model), different tests would be selected depending on the presentation or distribution of metastases, which would again lead to different outcomes
    • The number of possible subgroups to model was large, hence at the 7th GDG, agreement was reached to prioritise 2 subgroups for analysis: patients with liver metastases and patients with bone metastases.
At the 8th GDG, the revised draft model structures were discussed and there was consensus that the GDG should not proceed with the economic analysis for Topic 13. The following challenges contributed to this decision:
  • It was difficult for the GDG to agree on an exhaustive but mutually exclusive set of pathways in the economic model. For example, in clinical practice, after conducting PET-CT, the subsequent choice of confirmatory diagnostic test or treatment decision is influenced by a complex set of patient and disease factors. There were concerns that a more simplistic approach would not adequately reflect the scope of the decision problem faced by the clinician/patient.
  • There is limited data in the published literature to populate the economic model and some data would need to be elicited from experts, but the availability of experts on the GDG who have specific experience in each of these patient subgroups is limited.
  • A decision was made to focus on subgroups of CUP patients to develop a model that is clinically accurate. Each subgroup analysis would require considerable resource and time commitment from the health economists and GDG members to refine and seek agreement on the model structure and elicit data to populate the model. By focusing on subgroups, the results however would only be relevant for estimating the EVPI and informing the use of PET-CT in these subgroups of patients (i.e. a small proportion of CUP patients).
Topic 27Proposed analysis

Aim

An analysis will be carried out to assess the value of perfect information in a comparison of active chemotherapy vs best supportive care for the treatment of patients with confirmed cancer of unknown primary with no clinical features fitting a recognised confirmed CUP syndrome. The findings of this analysis will be used to inform future research recommendations.

Interventions
The most commonly used chemotherapy regimen(s) in current UK practice, to be defined with GDG input.

Comparators
Best supportive care, to be defined with GDG input.
Methods

Formal value of information analysis will provide an analytic framework which will address whether a decision regarding consideration of a systemic treatment can be made on the basis of current evidence or whether more evidence is required to support this decision about chemotherapy regimes in the future, and how much we should be prepared to pay for this evidence. (Claxton et al. 2002)

There are three components in this framework:
“(i) the construction of a decision analytic model to represent the decision problem; (ii) a probabilistic analysis of this model to characterize the current decision uncertainty; and (iii) establishing the value of additional information” (Briggs et al. 2006)

The structure of the analytic model will be informed by the data available in the literature and in consultation with the GDG.

The NHS perspective will be adopted; that is the health benefits and costs to be considered in the analysis will only be those relevant to the NHS. Relevant costs include those borne by Personal Social Services (PSS) as well as those that fall on the NHS itself. Unit costs will be derived from publicly available national sources whenever possible (e.g. NHS Reference Costs).

Probability distributions will be assigned to different clinical and cost parameters within the model so that a probabilistic sensitivity analysis can be carried out to assess the overall uncertainty of the model and the robustness of the results.

Feasibility issues

It is unlikely that the literature will be sufficient to populate all relevant parameters in the model. Expert elicitation will be considered where estimates from the literature are not available.

Key references

  • Bridgewater J, van Laar R, Floore A, Van'T Veer L. “Gene expression profiling may improve diagnosis in patients with carcinoma of unknown primary” British Journal of Cancer. 2008;98:1425–1430. [PMC free article: PMC2361712] [PubMed: 18414470]
  • Briggs A, Sculpher M, Claxton K. “Decision Modelling for Health Economic Evaluation”. Oxford University Press; 2006.
  • Buckhaults P, Zhang Z, Chen YC, Wang TL, St Croix B, Saha S, et al. “Identifying Tumor Origin Using a Gene Expression-based Classification Map” Cancer Research. 2003 Jul 15;63:4144–4149. [PubMed: 12874019]
  • Claxton K, Sculpher M, Drummond M. “A rational framework for decision making by the National Institute for Clinical Excellence” Lancet. 2002;360:711–715. [PubMed: 12241891]
  • Department of Health. “A Framework for the Development of Positron Emission Tomography (PET) Services in England”. London: Department of Health; 2005.
  • Fogarty GB, Peters LJ, Stewart J, Scott C, Rischin D, Hicks RJ. “The usefulness of fluorine 18-labelled deoxyglucose positron emission tomography in the investigation of patients with cervical lymphadenopathy from an unknown primary tumor” Head & Neck. 2003;25(2):138–45. [PubMed: 12509797]
  • Johansen J, Buus S, Loft A, Keiding S, Overgaard M, Hansen H, et al. “Prospective study of 18FDG-PET in the detection and management of patients with lymph node metastases to the neck from an unknown primary tumor”. Results from the DAHANCA-13 study” Head & Neck. 2008;30(4):471–8. [PubMed: 18023031]
  • Regelink G, Brouwer J, de Bree R, Pruim J, van der Laan BF, Vaalburg W, et al. “Detection of unknown primary tumours and distant metastases in patients with cervical metastases: value of FDG-PET versus conventional modalities” European Journal of Nuclear Medicine & Molecular Imaging. 2002;29(8):1024–30. [PubMed: 12173016]
  • Varadhachary G, Abbruzzese J, Lenzi R. “Diagnostic strategies for unknown primary cancer” Cancer. 2004;100(9):1776–85. [PubMed: 15112256]

Addenda to economic plan

The following substantive revisions to the plans set out in the section 3 above have been agreed.

DateQuestion number(s)Agreed change to number or type of analyses
30/04/2009Topic 13Agreement was reached not to proceed with economic model for this topic. An explanatory note will be included in the guideline.

Footnotes

a

This may be done by face-to-face meeting, teleconference or email as convenient

b

May be the project manager, a systematic reviewer or research fellow and/or the centre director or manager, as appropriate for the NCC and guideline

c

May be GDG chair, clinical lead and/or other members as appropriate

d

CCP Director or Associate Director who is taking the lead for the guideline

e

One of the CCP health economic Technical Advisors

Copyright © 2010, National Collaborating Centre for Cancer.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licenses issued by the Copyright Licensing Agency in the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

Bookshelf ID: NBK82150
PubReader format: click here to try

Views

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to pubmed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...