5.5 Clinical features and management 487

5.5 Clinical features and management 487

ESSENTIALS Patients with cancer may present after detection of an abnormality in a screening programme, with vague symptoms, or with advanced disease. The commonest symptoms are local, directly related to the anatomical position and size of the cancer. Systemic features are classically those of anorexia and weight loss, but there can be many other manifestations including fever, hypercalcaemia, and a range of paraneoplastic syndromes. Early diagnosis of cancer is rightly seen as an important route to improving the survival of patients. Imaging is the basis for all solid tumour diagnosis and staging, and biopsy remains the gold standard diagnostic tool. Molecular pathology has revolutionized our under- standing of the biology of cancer, enabled more accurate prognosis and prediction of response to therapy, and led the way in relation to the development of targeted therapies. Cancer comprises many different conditions with very different requirements for the basic elements of a population-​based cancer care service. The biggest change in the delivery of cancer manage- ment in the last 20 years has been the recognition that cancer man- agement is a multiprofessional and multidisciplinary effort. There is a hierarchy of aims for cancer treatment which stretch from cure to palliation. Four key measures to reduce the burden of cancer on the patient, their family, and on the healthcare system are: (1) early diag- nosis to maximize the chances of cure and minimize the morbidity of treatment; (2) access to best treatment irrespective of chronological age or home address of the patient; (3) increased physical activity; and (4) a ‘recovery package’ of care and support. Introduction Cancer is the commonest cause of death in the United Kingdom and by 2030 it is estimated that half the population will have had a diagnosis of cancer. The high prevalence of the disease explains why government and health systems pay so much attention to pre- vention, early diagnosis, treatment, and survivorship programmes. The complexity of cancer provision is not due to the numbers of patients who have (or potentially will have) cancer, but more to the simple fact that cancer is not a single disease: it is made up of many different conditions with, in some cases, very different requirements for the basic elements of a population-​based cancer care service. There is also the added complication that cancer often presents to doctors who only rarely encounter malignant disease. This problem paradoxically applies to primary care physicians, who see huge numbers of patients, only a small percentage of whom have cancer. The issue of a lack of familiarity with the diagnosis of a particular cancer and its management in primary care is further confounded by the diversity of malignant conditions. Presentation of cancer The diagnosis of cancer is arrived at in a variety of clinical scenarios. For example, patients may present after detection of an abnormality in a screening programme, or they may present with vague symp- toms to their primary care physician. A particular concern in the United Kingdom is that a high proportion of patients present in emergency departments (about 25%), and these patients often have advanced disease. Halfway along the continuum of screened and ad- vanced cases due to neglect of early symptoms, is a group of patients who have an incidental diagnosis. The incidental diagnosis of cancer is becoming increasingly fre- quent as the use of diagnostics such as CT and ultrasound is becoming routine in primary care and other specialties. Many cancers are now typically diagnosed as an incidental finding (e.g. kidney cancer). Although such diagnoses are to be welcomed as they often represent early stage and thus curable disease, they present the challenge of what might be appropriate treatment in the context of the concomitant dis- ease that was being investigated. For instance, an incidental small renal mass in the context of a patient with an aortic aneurism and cardiovascular comorbidities is a different therapeutic challenge to the presentation of a small renal mass in a young fit patient that has been diagnosed incidentally as a result of a CT being performed following a sporting injury. There is also the issue of whether incidental cancers have the same biology as those discovered as the result of the inves- tigation of a symptom. The treatment plan for an incidentally diag- nosed cancer requires knowledge of the natural history of the disease and the significance of the early stage of the cancer in question. 5.5 Clinical features and management Tim Eisen and Martin Gore† † It is with great regret that we report that Martin Gore died on 10 January, 2019.

488 SECTION 5  Principles of clinical oncology Localized symptoms of cancer The commonest symptoms of cancer often relate to its position and resulting local symptoms, with the consequences of a mass being directly related to its anatomical position and size. Local symptoms require a careful evaluation of their chronicity and associated symp- toms and signs, such as rectal bleeding, weight loss, or the finding of an abdominal mass in the case of intra-​abdominal malignancy. Pain Pain is classically due to compression by a mass on surrounding structures and can either result in a dull ache over the site or se- vere pain on movement. Some cancer pains are very difficult to deal with, such as the neuropathic pain due to nerve involvement, while others such as pain due to the involvement of a vertebra can often be palliated by a single fraction of radiotherapy. Long bone involve- ment must be diagnosed promptly to prevent pathological fracture, which is a serious complication because the morbidity of treatment and outcomes are significantly worse compared to intervention for a lytic metastasis with an intact cortex. Referred pain can cause confusion and requires the physician to be very aware of the diagnostic requirements when ruling out the presence of a referred pain rather than pain due to a local cause. Classic examples include shoulder tip pain from disease in the right upper quadrant of the abdomen, and pain and paraesthesia in limbs due to compression of nerve routes. Intra-​abdominal discomfort is perhaps the most difficult pain diagnostically, particularly in the context of primary care. Many pa- tients have symptoms of discomfort and bloating, which can be a classic symptom of intra-​abdominal malignancies such as ovarian cancer, but equally is common among the population because of the prevalence of conditions such as irritable bowel syndrome and diverticulitis. Headache and other symptoms Headache is a common presentation of many benign conditions and primary brain cancer is a rare condition: nevertheless, it is a very common presenting feature of tumour within the cranium and the ability to recognize which patients require a brain scan is important. Mild symptoms may be sinister, and filtering of symptoms is as- sisted by considering their context. For example, a hoarse voice or chronic worsening cough is particularly concerning in a smoker. Systemic features of cancer Systemic features of cancer are classically those of anorexia and weight loss, particularly in the case of upper gastrointestinal (GI) malignancies such as stomach and pancreatic cancer. There are, however, many other systemic features of cancer, some of which need to be carefully evaluated (e.g. fever and night sweats, or the concomitant symptoms associated with hypercalcaemia that can be difficult to recognize clinically unless the patient already has a malignant diagnosis). Hypercalcaemia has more than one genesis because it can occur as the result of both direct tumour destruction of bone, or result from ectopic secretion of parathyroid hormone or (perhaps more commonly) parathyroid hormone-​related protein. The presence of fatigue is common in cancer patients and can re- late to anaemia, but not invariably so, and measuring the haemo- globin level is not a screening test for malignancy. There is a very wide variety of so-​called paraneoplastic syndromes associated with cancer. These are usually associated with advanced disease and can be very challenging diagnostically outside of the better-​known syndromes such as dermatomyositis and clubbing. Endocrine, neurological, dermatological,
and haematological syndromes Endocrine disorders such as hypercalcaemia, hypoglycaemia, in- appropriate secretion of anti-​diuretic hormone and Cushing’s syn- drome are all associated with paraneoplasia. A variety of neurological presentations including cerebellar de- generation and peripheral neuropathy can also be associated with malignancy. Many of the paraneoplastic neurological syndromes are uncommon (e.g. dementia, transverse myelitis, limbic encephal- itis, optic neuritis, amyotrophic lateral sclerosis, and the myasthenic syndrome known as Eaton–​Lambert). There is often diagnostic delay, and such paraneoplastic phenomena often require very spe- cialist knowledge and management. An interesting feature of cere- bellar degeneration in some conditions such as ovarian cancer is that its onset can be associated with response to treatment rather than as an initial diagnostic presentation. There are many dermatological syndromes associated with malig- nancy, such as acanthosis nigricans, the syndrome of Leser–​Trélat (multiple seborrheic keratosis), erythema migrans, exfoliative dermatitis, panniculitis, porphyria cutanea tarda, and ichthyosis. Haematological manifestations other than anaemia include autoimmune haemolytic anaemia, microangiopathic haemolytic anaemia, thrombocytosis, granulocytosis, and erythrocytosis. However, perhaps the most important haematological paraneoplastic syndrome, because of its frequency, is that of thrombosis, which can be due to a combination of a hypercoagulative state and compression of veins, particularly in the case of intra-​abdominal cancers. Diagnosing and staging cancer Early diagnosis There is much attention paid to the early diagnosis of cancer as this is rightly seen as an important route to improving the survival of patients. The reason there is such attention paid to early diagnosis is that there is always a chance that early stage disease (of all tumour types) can be cured, whereas with advanced solid tumours cure rates are much lower. Early diagnosis is not simply an issue of diagnostic capacity re- lating to imaging and endoscopy, but rather one of public and healthcare worker education. Early symptoms of cancer need to be better understood by the population at large, and there is much dis- cussion as to how this might be best achieved. Many programmes tend to concentrate on one particular tumour type, such as cam- paigns relating to the early symptoms of lung cancer or colorectal cancer, but there also needs to be more attention paid to more gen- eral messages about health. Early diagnosis campaigns must be broad, including not simply high-​risk patients, but also programmes in schools and places of education.

5.5  Clinical features and management 489 Patients living in deprived areas and who have poor socio-​economic circumstance are particularly vulnerable to lower survival rates from cancer, and although there are many reasons for this, early diagnosis is one of the contributing factors, including poor health education and less access to healthcare and diagnostic provision. There is a tension between the cost-​effectiveness of screening pro- grammes and early diagnosis. The essential element, however, is that there should be ease of access to diagnostics in the primary care set- ting, so that the primary care physician is not faced with prioritiza- tion decisions relating to individual patients but can easily access those tests that they feel are required for a particular patient at a specific time. The biggest challenges relating to early diagnosis are the avail- ability of CT scanning and, perhaps most challengingly of all, en- doscopy. There are several ways in which CT and other imaging procedures can be expanded and made more available through technological advances, including remote access reporting, ex- tended working hours, and the provision through mobile units of scanning in work places and sites near to patients’ homes. However, the endoscopy challenge is much greater because the need is for very specific endoscopy suites and many highly skilled individ- uals to perform the procedure, both of which are space, time, and labour intensive. Investigation Imaging Imaging is the basis for all solid tumour diagnosis and staging. The ability to recognize the precise location of a local tumour and any metastases is central to prognosis as well as decisions relating to treatment. CT remains the most used and useful imaging technique in solid tumour oncology, and allows a fast and reliable way of obtaining an overall view of a patient’s cancer. The use of MRI has increased rapidly over the last 10 years, and as well as giving better definition (mainly for the treatment of individual disease sites, particularly in the primary setting), has given particular insights into the patterns of spread of disease. The use of diffusion weighted imaging has be- come a very useful tool in assessing peritoneal and bowel serosal involvement. Continuing developments in central nervous system (CNS) MRI have proven extremely useful when defining specific treatments for primary or secondary CNS tumours. Positron emission tomography (PET) scanning has opened up a new field of molecular imaging, which is allowing real time and in vivo characterization of tumours, particularly in relation to their response to novel therapeutics such as targeted agents. The ability of PET scanning to help define which masses contain active tu- mour and which do not has had very important beneficial conse- quences, particularly for malignancies such as germ cell tumours and lymphomas, where this information is now routinely required to plan ongoing management. It is in the area of molecular imaging that there is the most re- cent excitement. For instance, with novel immunotherapeutic ap- proaches it is becoming feasible by the use of specific probes to define T-​cell populations that are relevant to tumour response, ra- ther than having to biopsy patients. This work is still in its infancy and is likely to play an important role in the future design of novel targeted immunotherapies. Biopsy Biopsy remains the gold standard diagnostic tool, and advances in image-​guided biopsy (both in terms of the imaging hardware and bi- opsy instruments) have meant that diagnosis has routinely become an outpatient procedure with minimal discomfort to patients. Biopsy has also taken centre stage in terms of research due to the recognition that the knowledge of serial changes within the tumour may have particular prognostic and/​or predictive importance. There is thus a drive for ever less morbid biopsy techniques to allow for individual patients to have sequential samples taken. The knowledge of intrapatient changes, particularly related to a therapy, is of huge research value. So-​called liquid biopsies are also starting to come out of the ­research arena and enter into the realm of standard of care (see Chapter 3.10 for further discussion). We are on the cusp of a con- siderable change of direction in relation to biopsy material as we de- velop an increasing understanding of the significance of circulating tumour cells, DNA, and ribonucleic acid (RNA), and how this relates to the genomic heterogeneity known to exist within a single cancer in an individual person. The significance of the results obtained from liquid biopsy is variable. Our knowledge is more advanced in certain tumour types; for instance, work in prostate cancer has led the way with data suggesting that circulating tumour cells might be used to follow patients and their response to therapy. Serum tumour markers Serum tumour markers play an important role in diagnosis, assess- ment of treatment, and follow-​up of patients after initial therapy. There are few examples of tumour-​specificity relating to tumour markers, but there is no doubt that their measurement can play an important role in guiding diagnosis and helping evaluate the efficacy of subsequent treatment. Germ cell markers are the paradigm for the use of serum tumour markers, both in respect of specificity and following patients during and after treatment. However, even when tumour markers are not absolutely specific, the pattern of tumour markers can be useful (i.e. a comparison of the different levels of more than one marker in an individual patient at a single point in time). This is particularly true of intra-​abdominal malignancies, where the relative rise of CA125, CA19.9, and carcinoembryonic antigen (CEA) can guide diagnosis depending on the values of the different tumour markers in relation to one another. The tumour markers associated with solid tumours, however, have some important pitfalls, particularly in relation to follow-​up. If patients are followed-​up with tumour markers that often rise before there is any evidence of relapse on imaging, this can cause the pa- tient enormous anxiety. This is particularly problematic when there is no evidence that the early treatment of relapse is of benefit to the patient. The use of tumour markers in follow-​up for some cancers is therefore controversial, and it will remain so until new treatments are available that will benefit survival if instituted early at the first sign of relapse. There is little doubt that a serum tumour marker that appears very early in the disease and is specific would be an ideal tool for improving the mortality from cancer in a population, but it does not appear that such a simple solution is likely to exist in terms of a specific protein. However, perhaps the development circulating DNA and RNA technology may one day produce such tests.

490 SECTION 5  Principles of clinical oncology Molecular characterization of tumours Molecular pathology has revolutionized our understanding of the biology of cancer, enabled more accurate prognosis and prediction of response to therapy, and led the way in relation to the develop- ment of targeted therapies. The earliest example of molecular pathology impacting thera- peutics is in breast cancer, with recognition of the relationship be- tween the expression of oestrogen and progesterone receptors and response to hormone therapy. Breast cancer was also the first solid malignant disease to use molecular targeting, with the identification of HER2 and the development of Herceptin and other anti-​HER2 strategies. In solid tumour oncology the ability to define molecular drivers of disease, such as mutations in c-​kit in GIST, EGF receptor in lung cancer, and BRAF in melanoma, have had an enormous impact on the therapeutic landscape and outcome for many patients. Classic examples of the power of molecular pathology include the predictive value of NRAS mutations and the use of chemotherapy in colorectal cancer. In haemato-​oncology, the use of monoclonal antibodies such as rituximab and imatinib in lymphoma and chronic myeloid leu- kaemia, respectively. This ever-​expanding molecular knowledge and precision is not, however, without its difficulties. Definition of ever smaller patient population groups leads to problems in terms of evidence for the licensing of products, and equally (or perhaps even more import- antly) how those who pay for healthcare approach the cost effective- ness calculations of different targeted agents for cancer. Molecular pathology is moving us to a position where almost every patient within a tumour type may become a ‘molecular minority’ and that group may simply not be large enough for the traditional calcula- tions of treatment ‘worth’ to be made by the payers of healthcare. Thus, while molecular pathology has brought untold benefits to pa- tients, it is challenging our traditional ways of categorizing cancers, making therapeutic decisions and defining cost effectiveness. Staging cancer All patients diagnosed with cancer require ‘staging’. Staging is a con- vention whose definition and methodology of process is tumour-​type dependent. For instance, in some malignancies staging is defined at surgery, whereas for others stage is defined by imaging and other tests. Most cancers are now staged using the universal T (tumour), N (nodes), M (metastases) system, although some tumour types are still staged according to their own classification frequently denoted as Stage 1, 2, 3, and 4, with subcategories within each stage (e.g. Stage 1a, b, or c, and so on). Staging at surgery can also be subdivided into those patients where surgical stage is defined on macroscopic appearances at surgery, although more frequently the results of the pathology are also used to inform the final stage, with the prefix ‘p’ being used in front of the tumour–​nodes–​metastasis (TNM) stage to indicate that the pathology has been used to define the final stage. Staging has two purposes. The first is prognostication. Most def- initions of each separate stage within a system are based on retro- spective series where the prognosis is correlated with the criteria that define a particular stage +/​-​ substage. The second and more im- portant use for staging is to guide treatment, which is why accurate staging for each patient is paramount and an absolute requirement before therapy is instituted. Management Planning treatment Multidisciplinary teams The biggest change in the delivery of cancer management in the last 20 years has been the recognition that cancer management is a multiprofessional and multidisciplinary effort. Patients need to be put at the heart of multidisciplinary teams, each participant bringing to bear their own knowledge, expertise, and experience. There is no longer any place for the sole specialist to define how an individual patient is managed. Management and treatment pathways need to be predefined for different situations, and each patient’s case should be discussed in a multidisciplinary setting in order to set that patient’s needs in the context of known evidence-​based intervention. The multidisciplinary team for solid tumours typically consists of oncologists, surgeons, pathologists, radiologists, and specialist nurses, and in many incidences other healthcare professionals such as dieticians and speech/​language therapists. All are specialists in the particular tumour type(s) related to that team. The United Kingdom has probably got the most developed and systematic multidisciplinary healthcare framework in the world, such that it is extremely difficult for any cancer patient in the United Kingdom not to be discussed at such a meeting and their diagnosis and treatment plan recorded prospectively. It is therefore some- thing of a paradox that survival rates remain lower in the United Kingdom than many other parts of Europe, and this does call into question the relationship between the apparent benefits of multi- disciplinary working and the outcome for individual patients. Criticisms of multidisciplinary working, particularly in the United Kingdom, do not relate to the existence of such arrangements, but rather to the detail of how these arrangements are put into practice. Multidisciplinary meetings are lengthy, and there have been sugges- tions that those patients whose management is not controversial or in question need not be discussed in detail, so that colleagues can discuss more complex and difficult cases at greater length. There are also other benefits of multidisciplinary working and meetings, such as the development of a collegiate and cohesive team, particularly as it is recognized that functional relationships between colleagues leads to better outcomes for patients. The aims of treatment There is a hierarchy of aims for cancer treatment which stretch from cure to palliation. There has been a move to try and prospectively define for each patient the precise aim of treatment (i.e. cure, life prolongation, palliation), but these divisions for some solid tumours can be blurred and confusing. For instance, patients with metastatic disease can be cured, and depending on the disease in question the percentage of patients with curable cancer varies. Thus, in a group of patients it may be known that 20% will be cured, but it is rarely possible to prospectively identify which individual patients fall into this category. The patient, however, may have their life prolonged or their symptoms palliated. The treatment for that patient is the same even though there are three separate possible outcomes for them, none of which can be defined before treatment is instituted. The def- inition of the aims of a treatment thus need to be put into the context of probability, based on the knowledge of the pathology, molecular

5.5  Clinical features and management 491 definition, extent of disease, clinical prognostic factors of the pa- tient, and of course most importantly, the patient’s wishes. The term ‘curative intent’ is very often used, but this term might perhaps best be replaced by a more descriptive one such as ‘cura- tive potential’ as we develop more therapies that have the potential of prolonging life, but we do not yet know whether a truly cured state exists. The main issue, however, does not relate to semantic definitions but to careful explanations to patients of the different probabilities of various outcomes and, of course, the toxicity of treatment. The use of best supportive care strategies is paramount in those situ- ations where it is clear that life cannot be prolonged and cure is out of the question. There also needs to be more understanding that the knowledge and techniques used in best supportive care situations should also come into play at earlier stages of a patient’s treatment pathway. The early involvement of experts in palliative medicine has been shown to make a considerable difference to patient out- come and experience, and there are no discernible downsides to involving colleagues in palliative medicine in the management of patients going through curative, let alone palliative, therapies. Modalities of treatment There is an understanding that all modalities of treatment can be brought into play when managing patients. The use of chemotherapy and radiation together has been well described for many years, and how surgery relates to both these modalities has also been the subject of many randomized trials in many tumour types. There has been a change of philosophy recently in relation to the development of ablative therapies for patients with so-​called oligometastatic disease. The ablation therapies can be used in con- junction with systemic treatments, and although there are (as yet) no randomized data on this potential change in therapeutic strategy, it is undoubtedly going to be increasingly used for patients with relatively slowly progressing oligometastatic disease. This change of philosophy in relation to oncology strategies is related to both technology and biology, for instance, the increasing availability of stereotactic radiotherapy and its ease of use, our increasing know- ledge of the genomic heterogeneity of tumours, and in relation to immunotherapy the possibility that neo-​antigens can be exposed and created by both radiotherapy and physical ablative therapies such as cryotherapy or radiofrequency ablation. This is an exciting area of therapeutic strategy, not only because there is the theor- etical possibility of keeping disease under control for longer, but it suggests that if techniques are non​morbid and associated with little risk, then treating tumours when they are small before they can cause local or systemic symptoms becomes a logical approach to cancer management. Furthermore, if for instance two separate clones expand while others are kept under control by systemic therapy, then logic would dictate that these two clones should be ablated by a local therapy. Surgery Surgery undoubtedly cures more patients than any other modality of treatment and is the pivotal initial therapeutic intervention in solid tumour oncology. Other treatment modalities, such as radiotherapy and systemic therapy, are often built around surgery. For instance, treatment given after surgery—​whether radiotherapy or systemic therapy—​is described as adjuvant therapy, and sometimes both are used in this setting. Increasingly, systemic therapies—​particularly chemotherapy—​are being given before surgery in order to ‘down stage’ the tumour and allow surgery to take place more easily and with less morbidity. Treatment prior to surgery is known as neoadjuvant therapy and applies to either systemic therapies or to a combination of chemotherapy and radiotherapy (known as ‘chemorads’). Surgical techniques have improved remarkably in the past three decades. Use of minimally invasive techniques, such as laparoscopic surgery and robotically assisted surgery, have minimized the mor- bidity of interventions. An important use of surgery is in relapsed disease, and particularly in relapsed disease that is solitary and/​or slow-​growing. Frequently a single metastasis is observed for 3–​6 months and then removed if it remains solitary. This strategy is known as ‘a trial of time’. Molecular imaging techniques such as PET scanning are able to define single site disease with a good degree of accuracy and allow better patient selection for surgery at relapse. Common problems in cancer management Cerebral metastases The development of cerebral metastases is an extremely serious oc- currence in every tumour type; indeed, even in curable cancers such as leukaemia and lymphomas, particular attention is required to prevent CNS disease from occurring. Successful prophylaxis against CNS involvement was the reason why childhood leukaemia became such a curable condition. In solid tumour oncology the presence of solitary cerebral metas- tases is an indication for surgery. Surgery to such recurrences has a surprisingly low morbidity, depending on the site of the disease. Increasing use of stereotactic radiotherapy techniques have made a strategy of regular surveillance of the CNS an important area of ex- ploration because such techniques can destroy small tumours very effectively without morbidity. The argument remains as to whether or not such approaches are associated with increased survival. They are certainly associated with less morbidity than waiting for metas- tases to become large and/​or multiple for treatment to be instituted. Many stereotactic techniques can be used to treat more than one me- tastasis at any one time, or serial metastases that occur sequentially. The development of carcinomatosis meningitis, however, remains an extremely difficult complication to treat and in many tumour types is a preterminal event. Effusions The development of effusions in the serosal cavities of the body, namely the pleural, peritoneal, and pericardial spaces, are common complications of cancer. Symptoms include shortness of breath, car- diac tamponade, and abdominal discomfort. Patients require fluid to be drained so that their symptoms are palliated. It is often sensible to examine fluid drained from an effusion for infection (bacterial and TB) as well as the presence of malignant cells. In newly presenting patients, cytology can be extremely useful in making at least a preliminary diagnosis, and if cell blocks are used, considerable information can be gathered because such specimens are suitable for immunohistological examination and may facilitate accurate diagnosis.

492 SECTION 5  Principles of clinical oncology For patients in whom active therapy is no longer having an ef- fect against the disease, the control of effusions in the three serosal cavities becomes paramount for successful palliation. Permanent pleural drains can be used in both the pleural and peritoneal cavities, and pericardial windows can be created surgically for uncontrollable pericardial effusions without excessive morbidity for the patient. An important aspect of the treatment of pleural effusions is that of pleurodesis. It is greatly preferable that this is attempted under direct vision through a video-​assisted thoracic surgery (VATS) pro- cedure: using adhesive techniques through a simple chest drain may result in the patient developing loculated recurrent effusions, which are then very difficult to drain satisfactorily. Pleurodesis is only suc- cessful if the pleural cavity is drained to absolute dryness, and this can only be reliably accomplished under direct vision. Oncological emergencies The term oncological emergency requires precise definition because there are situations when intervention is required as an absolute im- mediate need, a true emergency, and others where treatment needs to be urgently instituted, within 24–​48 hours. These latter clinical presentations are seen with uncontrolled, undiagnosed disease and include cachexia, anaemia, and those relating to the side effects of treatment such as nausea and vomiting. True emergencies (i.e. those that require immediate and specific management) are in fact few and far between in oncology. The two main true oncological emergencies are neutropenic sepsis and cord compression. Superior venocaval obstruction is sometimes de- scribed as the third oncological emergency, but it rarely requires intervention within hours, although rapid intervention is desirable because patients can be extremely distressed and uncomfortable when this complication of cancer occurs. Neutropenic sepsis The survival of patients from neutropenic sepsis is directly related to the time between patient presentation and the institution of intra- venous antibiotics, and there has quite correctly been considerable emphasis placed on the importance of there being as short a time as possible between patients presenting with neutropenic sepsis and the institution of treatment. The best way of ensuring that patients do not die from neutropenic sepsis is to make sure that both pa- tients and healthcare workers, particularly in emergency rooms, are aware of this complication. The prophylactic use of granulocyte-​ colony stimulating factor has made an enormous difference to the incidence of neutropenic sepsis and in some healthcare economies it is used routinely, thus making this oncological emergency a rarity. Cord compression Management of patients with cord compression remains a challenge. Neurosurgical services may find it difficult to respond in a timely manner to requests for assessment because of the other demands that they face. Carefully designed networks and pathways for patients with suspected cord compression are paramount, and these should be en- gaged each time there is a suspicion of cord compression. There also needs to be a better understanding that the treatment of cord com- pression is best instituted when there is impending cord compres- sion, rather than waiting for overt physical signs of paralysis which inevitably results in a delay to recovery. Networked arrangements for the management of cord compression should include the efficient and timely availability of specific imaging, and assessment of patients by both radiation oncologists and neurological or spinal surgeons. Often the decision to engage radiotherapy or surgery is straightfor- ward, but in patients who have other sites of disease or perhaps more than one vertebral site of involvement, the decision between radio- therapy and surgery can be difficult and complex. Add to this the time-​critical nature of the correct treatment being instituted, and the seamless networking of diagnosis and decision-​making become one of the most important aspects of creating a functional cancer service. Other emergencies Rare emergencies such as hypophysitis, adrenal insufficiency, and insulin-​dependent diabetes have emerged with the development of novel immunotherapies, and while the management of these is exactly the same as for their sporadic counterparts, their recognition may become an increasingly important issue as these agents become more frequently used. Cancer in particular patient groups The frail Cancer is often described as a disease of old age, with many solid tumours having their median age of onset during the sixth and later decades (see Chapter 5.1). This fact, together with the increasing in- cidence of cancer and age of the general population, means that more and more patients who are frail and have multiple comorbidities are going to be presenting to cancer services. Decisions about the appropriateness of both diagnosis and treat- ment are becoming ever more complicated, and while there is an im- portant drive to prevent discrimination in relation to the availability of cancer treatments based on age, there is also the tension between the ability to deliver effective cancer treatment in frail patients and the de- sirability to do so. We know that there has been discrimination against elderly patients in relation to the delivery of cancer treatments in that older patients are less likely to be offered a standard of care compar- able to that of younger patients, but to leap to the conclusion that all elderly patients should be offered all standard treatments is to deny the nuances of comorbidities as they relate to older people. Much needed attention is now being given to this area, and this is starting to help us make treatment decisions for older people and, in particular, the challenge of those frail elderly patients who have reduced cognition. There are processes in place to protect such vulnerable adults from therapeutic discrimination. Nevertheless, there are still considerable social challenges relating to the care of the frail and elderly, as well as financial and physical ones. The area requires increased efforts in rela- tion to research and as healthcare systems are becoming increasingly stretched, it is encouraging that such research is becoming recognized by governments as something of an imperative. In pregnancy Cancer in pregnancy used to be thought of as an absolute indication to terminate the pregnancy if at all possible. However, in more recent times there has been increasing evidence that systemic therapies and surgery can be safely used in cancer patients who are also pregnant. Typically, surgery is safe in the early stages of pregnancy, while sys- temic therapies are safe after the first and early second trimester. There are some typical cancers that occur during pregnancy based on age, and these include the lymphomas, leukaemia, melanoma,

5.5  Clinical features and management 493 and germ cell tumours. The important strategic thinking for such patients is to design the therapeutic approach as if the patient were not pregnant, and then see if it is applicable to that particular stage of pregnancy. It is becoming less common to recommend termination to pregnant women with a concomitant cancer, although occasion- ally this is necessary. See Chapter 14.18 for further discussion. Hereditary cancer There is increasingly knowledge of the familial nature of some can- cers and an increasing knowledge of how patients are managed in this situation. In particular, knowledge of potential second or third malignancies associated with a genetic mutation in a cancer patient is becoming relevant as more primary tumours are becoming poten- tially curable. A classic example of this are BRCA mutations associ- ated with breast, ovarian, and prostate cancer. Where multiple cancers may occur with a hereditable mutation, screening may become complex, for instance, in patients with Lynch syndrome where very specific expertise is required to advise patients on how they should be managed following treatment of their pre- senting tumour. Cancer genetics goes wider than the individual patient and in- cludes the family. Family dynamics can be severely altered by the knowledge that a family member carries a potential hereditable mu- tation, and considerable care needs to be taken to treat each indi- vidual family member as such, an individual who requires separate and specific advice as it relates to them. It is not uncommon when a hereditable mutation is discovered in a family that the family attempt to take a ‘unified’ family approach, but this can be very damaging and the confidential provision of advice for individual family mem- bers is paramount. In conditions such as breast and ovarian cancer, where knowledge of a hereditable genetic mutation may impact the patient’s treatment (e.g. poly ADP ribose polymerase (PARP) inhibitors and BRCA-​ mutated ovarian cancer), there can be unforeseen consequences of wishing to detect such mutations in the patient who already has cancer. For instance, a daughter may then discover that she could potentially have inherited a high risk of breast or ovarian cancer and may be severely disturbed by such knowledge. Similarly, there may be an assumption from their mother that they would wish to know this so that prophylactic measures can be taken, but this can imme- diately bring mother and daughter into conflict. Experience, how- ever, suggests that these conflicts tend to be few and far between, and programmes that routinely screen patients for BRCA mutations have not reported detriment in terms of the psychodynamics within families. Information and support for patients and carers At diagnosis Many patients will have definite expectations of what a cancer diag- nosis will mean in terms of treatment, survival, and long-​term ef- fects. The accuracy of these expectations will depend on many factors, such as past experiences with friends or family members in a similar situation, information from news or other media. It is extremely important to communicate with cancer patients, their family, and friends, in an honest, open, and sensitive way. Important elements of successful communication include: • The patient will very often feel extremely vulnerable, and the sup- port of a family member or friend can be invaluable. • The meeting should take place in a quiet environment without interruption. • The doctor, clinical nurse specialist, and other staff involved should all be familiar with the patient’s case and the manage- ment plan. • Providing ‘chunks’ of information rather than a single lengthy nar- rative. Similarly, it may be helpful to go over information on sev- eral occasions. • Written information sheets are an invaluable aide-​mémoire and should complement rather than replace personal communication. High-​quality information sheets are available on a very wide range of subjects from major cancer charities, with smaller specialist or local charities providing tailored support for the precise condition or locality. • The clinical team should provide a clear description of the patient’s treatment options, including both standard of care and research options. Noting issues discussed earlier in this chapter, this de- scription should include the purpose of treatment—​whether it is to palliate symptoms, prolong life, or to provide a cure. Unless starting treatment is an emergency, it is considered best practice to ask the patient to consider these options for a period of a few days. Further questions may arise and it is helpful to ask the patient to write them down so that the clinical team can answer them when the patient returns to clinic. • Patients should be fully involved in making decisions. Sensitivity and common sense are required in balancing an unbiased discus- sion of options with the patient’s need to make an important deci- sion in an unfamiliar setting. It is entirely reasonable and indeed desirable for the clinical team to provide clear guidance. At different stages of the cancer journey Macmillan Cancer Support worked with the National Cancer Intelligence Network to provide a detailed survey of healthcare and outcomes for 85 000 patients in England and Wales between 2004 and 2011. This strongly suggested the need for four key measures to reduce the burden of cancer on the patient, their family, and on the health system: • Early diagnosis to maximize the chances of cure and minimize the morbidity of treatment; • Access to best treatment irrespective of chronological age or home address of the patient; • Increased physical activity; and • A ‘recovery package’ of care and support. Patients need different types and levels of support at different points during their cancer journey. Maher and McConnell estimated the number of people in six distinct periods of their cancer journey: (1) diagnosis and treatment; (2)  rehabilitation; (3)  early monitoring up to 5 years, and (4) up to 10 years; (5) progressive disease; and (6) end-​of-​life care (Fig. 5.5.1). They found that the greatest need for

494 SECTION 5  Principles of clinical oncology support was at times of transition. Most clinicians would recognize the need for support during the transition from healthy to diagnosis and treatment, and also from a period of monitoring to diagnosis of relapse and the need for further treatment. Fewer will be so aware of the need for support during transition from treatment to a period of rehabilitation. Understanding the individual needs of patients and their fam- ilies, particularly at these periods of transition, will enable health staff to provide necessary targeted support at the optimal time. The goal of empowering patients to deal with the medical, psycho- logical, social, and financial impact of cancer diagnosis and treat- ment is key. Diagnosis and treatment 41000 Diagnosis and treatment 40000 Rehabilitation 12000 Rehabilitation 28 000 Early monitoring: 2 ≥ 5 years 9000 Lung cancer (b) (c) Early monitoring: 2 ≥ 5 years 45 000 Colorectal cancer Early monitoring: 5 ≥ 10 years 6000 Early monitoring: 5 ≥ 10 years 51 000 Later monitoring 21000 Later monitoring 73000 Progressive illness ??? Progressive illness 24000 End-of-life care 35000 (28 000 year 1) End-of-life care 16000 (11 000 year 1) Diagnosis and treatment 48000 Rehabilitation 44000 (a) Early monitoring: 2 ≥ 5 years 100000 Breast cancer Early monitoring: 5 ≥ 10 years 122000 Later monitoring 226000 Progressive illness 24000 End-of-life care 12000 (2000 year 1) Fig. 5.5.1  Periods of the cancer journey. (a) Breast cancer care pathway: estimating the number of women in the United Kingdom, 2008*. (b) Colorectal cancer care pathway: estimating the number of people in the United Kingdom, 2008*. (c) Lung cancer care pathway: estimating the number of people in the United Kingdom, 2008*.

• For each cancer type, the size of the boxes reflects the approximate proportion of people in each phase (with double counting for people who are diagnosed and die in the same year—​these numbers are indicated in brackets; i.e., ‘XX 000 year 1’). Median survival for incurable disease was taken from Frontier Economics (2010) ‘One to one support for cancer patients: a report prepared for Department of Health’ and is 3 years for breast cancer and 2.5 years for colorectal cancer. Estimates for progressive illness for lung cancer have not been made. Estimates for later monitoring for lung cancer exclude 8000 men more than 20 years from diagnosis. The total for men in this group was thought to be an overestimate and is likely to be nearer 6000 than the modelled 14 000. Various sources including Office for National Statistics, Cancer Research UK, Frontier Economics, Information Services Division (ISD) Scotland, Northern Ireland Cancer Registry, Welsh Cancer Intelligence and Surveillance Unit, London School of Hygiene and Tropical Medicine. Reprinted by permission from Macmillan Publishers Ltd on behalf of Cancer Research UK: British Journal of Cancer (Maher J and McConnell H, 2011. New pathways of care for cancer survivors: adding the numbers. Br J Cancer, 105, S5–​S10), copyright © 2011.

5.5  Clinical features and management 495 While it is normal for patients to feel very upset by the news of a cancer diagnosis, it is important to recognize when psychological support is required. This may not be obvious and a holistic assess- ment of care needs early in the patient’s cancer journey often reveals unexpected problems and fears. Helping a patient to deal with these can significantly boost quality of life under difficult circumstances. Many cancer units offer specialized psychological support services. For patients of working age, the ability to return to work is an extremely important element of rehabilitation. Evidence from the United Kingdom, the United States, and the Netherlands suggests that the impact of cancer on employment status is transient for most people, with around 80% able to return to work eventually, the corollary of which is that around 20% of people do not return to work. Four key factors determine the ability of somebody to return to work after a cancer diagnosis: • The organ of origin of the cancer: patients with lung cancer or myeloma were much more likely not to return to work than pa- tients with breast or urogenital tumours; • The intensity of treatment: patients presenting with late stage dis- ease requiring intensive treatment, or with tumours requiring aggressive therapies, are much more likely to have difficulties re- turning to work; • Occupational status: cancer survivors with physically demanding jobs are 20% less likely to be in employment 2–​3 years after diag- nosis than age-​matched controls. In contrast, those with sedentary occupations are only 7% less likely to be in employment; • The role of others in facilitating a return to work: employment protection and anti-​discrimination legislation has greatly im- proved the position of cancer survivors wishing to return to work since the 1970s. After cancer Survivorship In 1970, the average life expectancy after a diagnosis of cancer was one year. By the 1990s this had doubled. Due to a combination of earlier diagnosis and more effective anti-​cancer treatments, the average life expectancy for a patient diagnosed with cancer was close to 10 years in England and Wales in 2010–​11. Cancer survival in several other advanced health economies is even more impressive. Due to these improvements in survival, the number of people living with or after cancer has increased enormously. For example, in the United Kingdom 1.2 million were living with or after a diagnosis of cancer in the early 1990s, rising to 2.5 million by 2015. The number is projected to rise to 4 million in 2030. In themselves, the survival figures are most encouraging, but these crude statistics conceal a much more complex and difficult picture for patients with cancer. The Macmillan Cancer Support/​ National Cancer Intelligence Network survey confirmed several well-​recognized differences in clinical outcome between four cancer types: breast, prostate, lung, and brain/​central nervous system ma- lignancies. The main impact of the survey, however, was to reveal Table 5.5.1  Long-​term sequelae of cancer treatment Consequence of treatment Notes Growth abnormalities Radiation involving a growing bone epiphysis may halt growth locally and result in unilateral shortening. Similarly, steroids and chemotherapy drugs can impair growth. Reduced fertility In men, rapid loss of fertility commonly follows treatment with alkylating agents or irradiation of the gonads. Sperm storage is essential before therapy begins. Some men will recover their fertility up to 2 years after treatment is completed. In women, chemotherapy and radiotherapy of the pelvis can lead to premature ovarian failure, especially high-​dose treatments combining irradiation with alkylating agents. Strategies to maintain fertility should be implemented before therapy starts. The strategies may include ovarian suppression during therapy and, for women not requiring urgent cancer treatment, storage of frozen oocyte or embryo. In women who retain fertility, it is likely that menopause will occur some years earlier than would otherwise have been the case. Cognitive impairment Treatment of central nervous system tumours, especially in the young, may result in cognitive impairment with profound consequences for the patient and their family. Renal impairment Treatment with regimens involving cisplatin or ifosfamide may result in renal tubular damage and reduced glomerular filtration. Cardiac impairment Irradiation of the mediastinum and left chest wall, as well as the use of anthracyclines or vincristine can lead to cardiac toxicity including cardiac failure, arrhythmias, and risk of myocardial infarction. In adults, limiting the total dose of anthracyclines reduces the risk of cardiac toxicity. Peripheral neurological impairment Many chemotherapy agents are neurotoxic with cisplatin and vinca alkaloids being the best recognized. Recognition of this toxicity during treatment may lead to modification of dose or regimen. Respiratory impairment Fibrosis may result from radiotherapy or multiple drugs, notably bleomycin. Similarly, pneumonitis is a well-​recognized complication of many therapies and may respond to dose interruption with steroids. Gastrointestinal impairment There are multiple causes of dysfunction, including pelvic irradiation. These symptoms need expert assessment. Immune dysfunction With the development of T-​cell checkpoint inhibitors, we are noticing autoimmune phenomena, which may continue, or rarely start, after cessation of treatment. These toxicities include rash, colitis, endocrine abnormalities, and pneumonitis. Osteoporosis Chemotherapy, radiotherapy, and hormonal therapy such as aromatase inhibitors may result in osteopenia. Bisphosphonate treatment and calcium supplementation may be required to maintain bony integrity. In extreme cases, high-​dose steroids may lead to osteonecrosis of weight-​bearing bones. Young patients may require serial joint replacements. Second cancers Both radiotherapy and several chemotherapies increase the risk of second malignancies. Tumours may arise either within or on the edge of the radiation field, often with a delay of 10–​30 years.

496 SECTION 5  Principles of clinical oncology a far less well-​recognized issue, namely the extent to which people living with or after a diagnosis of cancer experience a range of other health and social conditions, limiting both the quality and the length of their lives, and having a major impact on their carers and children. To take breast cancer as an example, most (69%) women who re- ceive a diagnosis of breast cancer will survive at least seven years. Over a quarter of these women will have or will develop one or more serious comorbidities other than a relapse of their breast cancer. These conditions may both reduce the quality of life for the patient and make it more difficult to provide treatment. When compared with patients receiving hospital treatment for non​malignant con­ ditions, cancer patients have a much higher risk of comorbidities. For those patients with breast cancer who do not survive seven years after diagnosis, the risks are substantially higher, with 67% of all people with breast cancer having or developing one or more serious health conditions. Late sequelae of treatment The main long-​term sequelae of cancer treatment are noted in Table 5.5.1. With greater success in treating cancers, there is now greater realization of the problems that can be induced by current cancer therapies. These may occur in children and young adults through the effect of cancer and its treatment on the well-​being and development of maturing people. Equally, with the successful treatment of older patients with multiple medical problems, there may be exacerbation of underlying non​malignant conditions. As new therapies are introduced, clinicians need to remain vigilant as to their long-​term side effects. Since cancer survivors are at increased risk of many malignant and non​malignant conditions, it is important that they are encour- aged to have a healthy lifestyle by maintaining a normal weight, taking exercise, and avoiding smoking and excessive alcohol. FURTHER READING Maddams J, et al. (2009). Cancer prevalence in the United Kingdom: estimates for 2008. Br J Cancer, 101, 541–​7. Maher J, McConnell H (2011). New pathways of care for cancer sur- vivors: adding the numbers. Br J Cancer, 105, S5–​S10.