Palliative care and quality of life

Quality of life in ovarian cancer

Because the majority of patients with ovarian cancer present in advanced stages, most women enter treatment with some QoL dysfunction likely related to symptom severity at presentation. The study of QoL in ovarian cancer therefore must take into consideration this baseline level of dysfunction and monitor the effect of surgery or chemotherapy. Because the prognosis is generally poor for those with advanced ovarian cancer, the goal of therapy is perhaps to extend life without greatly compromising the patient’s QoL. In clinical trials, great attention is now paid to measuring QoL throughout the study of cancer therapeutics. QoL in ovarian cancer has been described to include three specific groups: (1) Patient undergoing maintenance therapy in a low-risk situation, (2) patient undergoing maintenance therapy in a high risk for recurrence situation, and (3) patient receiving therapy for active disease. QoL can be monitored posttherapy in survivorship/remission, to document the needs and deficiencies of these women. In the measurement of QoL, a standardized and validated measure, the Functional Assessment of Cancer Therapy-Ovarian (FACT-O), is frequently used in clinical trials. The FACT-O is a combination of the FACT-G (General) plus additional questions related specifically to ovarian cancer ( Tables 19.3 and 19.4 ). The FACT-G is composed of four subscales: the physical, social/family, emotional, and functional well-being. Patients can answer on a scale of 0 to 4, with 0 being not at all and 4 being very much. In the physical well-being subscale, the patient is asked to elaborate on pain, nausea, and lack of energy and how these physical symptoms affect their life and how bothered by these symptoms the patient is. The social/family well-being subscale addresses the amount of emotional support the patient has from her friends, family, and partner. The emotional well-being subscale targets the patient’s feelings and emotions regarding her illness and how anxious she is about dying from this disease. Finally, the functional well-being subscale concerns the patient’s ability to function in society, such as her ability to work, sleep, and enjoy life. This subscale looks at the patient’s ability to accept illness and move on to live a “normal” life again. In the final section of the FACT-O, there is an additional concerns scale that targets specific symptoms, life-altering changes, and chemotherapy in ovarian cancer. In 2011, Donovan et al. performed a comprehensive review of the QoL literature related to ovarian cancer between the years 2000 and 2011. They sought to find the most important symptoms that should be recorded to measure QoL in ovarian cancer trials. The panel of experts identified the following “core” ovarian cancer symptoms: abdominal pain, bloating, cramping, fear of recurrence or disease progression, indigestion, sexual dysfunction, vomiting, weight gain, and weight loss. In addition, ovarian cancer–specific symptoms were suggested to be of value in the design of future QoL endpoints: abdominal pain, bloating, cramping, fear of recurrence, indigestion, sexual dysfunction, vomiting, weight gain, and weight loss. Now, with the complexity of the different situations for an ovarian cancer patient, the tools used to measure QoL can vary and may include composite measures such as the Quality-adjusted Time Without Symptoms of disease progression or Toxicity of treatment (Q-TWiST) which takes into consideration quality-adjusted progression-free survival (PFS). Depending on the patient situation, the goals of therapy consider the potential for either early or late relapse, and the control of secondary side effects balanced with a patient’s hope for a successful therapy.

The tools used to measure QoL involve the European Organization for Research Treatment of Cancer (EORTC) Core Quality of Life questionnaire (EORCC QLQ-C30). There is also an EORTC Ovarian Cancer Module (OV28). Most prospective randomized trials in the United States have used the FACT-O measure to assess QoL. The Gynecologic Oncology Group (GOG) and now NRG Oncology have used the FACT-O in prospective phase III randomized trials to monitor QoL during various experimental treatment protocols ( Table 19.5 ). The first ovarian cancer GOG trial using the FACT-O was published by Wenzel et al. in 2005. They published the QoL results for a randomized trial of interval secondary cytoreduction in advanced ovarian cancer patients. In this trial (GOG protocol 152), 424 patients were enrolled, and 380 (90%) of these women completed the initial and midtreatment questionnaires. Roughly 80% of the patients continued to complete the measures at the second through fourth assessment points. In this study, a better QoL as measured by the FACT-O was associated with improved overall survival (OS). Although the FACT-O results did not differ between the groups receiving this surgical intervention or not, this trial did emphasize that improvements in QoL existed for women in both arms at 6 and 12 months after therapy.

Notably, an effective treatment regimen demonstrated for women with stage III ovarian cancer that has been optimally surgically debulked is a combination of intraperitoneal (IP) and intravenous (IV) chemotherapy. GOG 172, a phase III trial comparing IP cisplatin and paclitaxel plus IV paclitaxel with IV cisplatin and IV paclitaxel, demonstrated that the IP/IV regimen was associated with the longest median survival time (65.6 months for IP/IV vs. 49.7 months for IV) that has yet been reported in women with optimally debulked stage III ovarian cancer. This statistically significant and clinically meaningful difference between IP and IV chemotherapy prompted a National Cancer Institute (NCI) Clinical Announcement (2006) alerting the public and health professionals to the superiority of IP chemotherapy in the optimal disease setting. However, the superior survival outcomes, demonstrated with the IP regimen in GOG 172, was associated with considerable toxicity, and patients randomized to IP therapy reported significantly worse QoL before cycle 4 ( P < .0001) and at 3 to 6 weeks after treatment ( P = .0035). However, there were no significant overall QoL differences between regimens 1 year posttreatment. Despite the reported benefits of IP therapy for PFS and OS, this study demonstrated physical and functional well-being deficits both during and after therapy in the IP arm. Neurotoxicity and abdominal discomfort were also more prevalent in the IP arm. Although the QoL scores improved over time, specifically at 12 months from initial treatment, this manuscript perhaps prompted physicians to reconsider this treatment modality secondary to toxicity concerns. In addition, IP is a more difficult treatment for the treatment team to administer as well as potentially more toxic. However, ultimately, the QoL differences between regimens were not significant long term. In 2015, Wright et al. looked at the use and effectiveness of IP chemotherapy in 823 women at six different institutions. These authors demonstrate a relatively poor acceptance of IP chemotherapy (<50%) despite published survival advantages. They cite certain factors as potential barriers to IP/IV chemotherapy usage including toxicities, patients’ preferences, the inconvenience of administration, potentially higher rates of extra-abdominal cancer recurrences, and beliefs that other chemotherapy regimens may be comparable with fewer toxicities. This study did demonstrate improved survival advantages, but IP chemotherapy usage is associated with younger age and fewer medical comorbidities, thus perhaps explaining the less than 50% adoption of this route of treatment.

In 2009, von Gruenigan et al. combined the results from the previously-described trials and reported the specific line items in the FACT-O that were reported between cycles. For example, the authors report significant differences in concerns such as pain, nausea, feeling ill, and side effects of treatment. Because significant functional well-being deficits were also reported, they suggest interventions aimed at specific side effects as a means to improve patients’ functional status. Social well-being went relatively unharmed except for satisfaction with one’s sex life. Emotional deficits were also noted across all items except for acceptance of illness. This study demonstrated that when broken down into individual line items, perhaps interventions can be designed to target improvement in these various areas. In 2013, Monk et al. reported on the QoL outcomes of a randomized trial incorporating bevacizumab into the carboplatin and paclitaxel IV chemotherapy. This study included a maintenance of bevacizumab versus placebo for a total of 22 cycles (treatment given every 3 weeks). During the initial chemotherapy phase of six cycles of carboplatin and paclitaxel, the patients on bevacizumab experienced significantly lower QoL scores (worse QoL) than those on chemotherapy alone. It was curious that bevacizumab compromised QoL during the chemotherapy phase only of treatment. Most striking, though, is perhaps the idea that extending PFS did not translate into enhanced QoL according to the authors. However, although QoL was not improved, the authors use the QoL data to conclude that this new regimen is an acceptable alternative to standard care.

Several European studies have also evaluated QoL in prospective trials. Although these studies use different QoL tools, it is important to comment on some of these studies in which QoL data in the United States are lacking. First, in 2010, Vergote et al. published the results of a randomized trial of neoadjuvant chemotherapy followed by interval surgical cytoreduction versus primary surgical debulking followed by chemotherapy. The QoL results were then published in 2013. Per the authors, QoL was assessed by the EORTC Quality of Life Questionnaire Module-C30 (QLQ-C30) core questionnaire version 3, which consists of a global health status and QoL scale, multi-item functional scales and symptom scales, and a scale regarding financial difficulties. The trial results demonstrated similar OS outcomes between the two groups. The authors chose to limit inclusion to the QoL analysis to institutions with at least 50% compliance and more than at least 35% during the follow-up period. There were no statistically significant differences between the two groups, although those with the primary debulking group at baseline had more pain but less dyspnea.

In addition, the Avastin Use in Platinum-Resistant Epithelial Ovarian Cancer (AURELIA) trial studies the use of physician’s choice chemotherapy with or without bevacizumab in platinum-resistant recurrent ovarian cancer. Using the EORTC Quality of Life Questionnaire–Ovarian Cancer Module 28 (EORTC QLQ-OV28) and Functional Assessment of Cancer Therapy–Ovarian Cancer symptom index (FOSI), the investigators demonstrated that with the use of bevacizumab, there was greater than 15% improvement in QoL and abdominal or gastrointestinal (GI) symptoms specifically. The CALYPSO trial was another European trial that randomized patients with platinum-sensitive ovarian cancer to carboplatin and pegylated liposomal doxorubicin versus carboplatin and paclitaxel. The QoL results, published in 2012, were generated by the EORTC QoL-QC30 questionnaire and OV28 ovarian cancer module. In this trial, patients treated with carboplatin and pegylated liposomal doxorubicin had improved QoL scores and fewer cancer- and chemotherapy-related symptoms. This study, although designed as a noninferiority trial, did suggest that improved outcomes with the experimental arm were not at the expense of decreased QoL.

More recently several randomized trials have described QoL endpoints including final QoL analyses for GOG213, GOG252, GOG258, and several phase III trials in the maintenance treatment setting. In addition, now several other versions of the FACT-O are acceptable such as the FOSI and FACT-O Treatment Outcome Index (TOI). Some noteworthy trials over the last several years include GOG213 which specifically introduced the exploration into the effects of a secondary surgical cytoreduction on QoL changes. Using the FACT-O-TOI, the authors indicate that although QoL may be worse after surgery, the QoL scores ultimately match between the surgery and non-surgery arms at 6 weeks and beyond.

Most recently, data has been published on patient-reported QoL (PRQOL) during maintenance therapy in ovarian cancer patients. These data are confusing, likely due to the timing of the measurements and the lack of the ability of the maintenance arm to demonstrate improvement in PRQOL over placebo. Some believe that without QoL improvements, maintenance therapy should not be administered unless survival outcomes prove beneficial. Despite the limitations in QOL assessment, the trials in maintenance therapy with poly (ADP-ribose) polymerase (PARP) inhibition did not demonstrate differences between the control and treatment arms, meaning introduction of maintenance therapy did not seem to introduce PRQOL detriments. Furthermore, the introduction of the Q-TWiST analysis helped to describe QoL changes in a maintenance setting. Q-TWIST incorporates both quality and quantity of life. QAPFS is quality-adjusted PFS and includes the importance that a patient places on their clinical state. TWiST is the period without clinically significant symptoms after randomization and before disease progression. In multiple studies the mean duration of TWiST and the QAPFS both favored the treatment arms utilizing PARP inhibitor therapy. The significance of this is that the toxicity of the therapy is not demonstrating a detrimental impact in the improvement of PFS.

Quality of life in cervical cancer

Although patients with cervical cancer may present with pelvic pain or bleeding (or both) in more advanced stages, cervical cancer may also be detected based on screening tests alone. The usual treatment involves surgery for early stage followed by possible radiation, chemotherapy, or both for high-risk cases versus chemotherapy and radiation alone for more advanced stages. Patients with cervical cancer present with a unique set of symptoms, side effects from treatment, and socioeconomic issues not seen in patients with ovarian cancer. For example, these women have a lower median age at presentation; this group also has a larger percentage of Latina or non-White patients and a larger percentage of lower-income patients. Furthermore, the chemotherapy and specifically the radiation received by these women produce such symptoms as sexual dysfunction and urinary and bowel dysfunction that perhaps affect women in unique ways.

In the measurement of QoL for cervical cancer, there are, similar to ovarian cancer, two tools that are most widely used, the FACT-Cx and the QLQ-CX24. The FACT-Cx is composed of the same physical, social, functional, and emotional well-being subscales as the FACT-G but in addition it has a seven-item scale for concerns related specifically to cervical cancer. These concerns involve vaginal discharge, bleeding, odor, narrowing or shortening, constipation, and dysuria. The GOG has published studies in advanced cervical cancer that have used the FACT-Cx. These patients have a poor median survival time, typically less than 6 months, and therefore it is still of debate whether or not treatment is futile in this setting. This is especially of concern in such advanced cancer when symptomatology can be debilitating and further aggressive treatment perhaps does more harm than good. This is an ideal patient population to study QoL because if QoL decreases with treatment, then one might argue against treatment in a disease in which OS is so poor.

In 2006, GOG protocol 169 was published by McQuellon et al. and reported QoL endpoints in a randomized trial that compared cisplatin versus cisplatin plus paclitaxel in advanced cervical cancer (see Table 19.5 ). Despite the impressive deficits in QoL compared with the general population, these two treatment regimens did not differ with regard to QoL. The cisplatin and paclitaxel arms did demonstrate improvements in response rate and PFS; therefore, the QoL data help to promote this regimen. In 2005, the QoL data from GOG 179 were published by Monk et al. This randomized trial studied cisplatin as compared with cisplatin and topotecan in advanced or recurrent cervical cancer cases. The FACT-Cx was used to document QoL between these two regimens in addition to exploring its relation to prognosis. There were statistical differences in QoL between the two regimens, but of interest was the relationship of the FACT-Cx to OS. As the FACT-Cx score increased, so did OS, and this was statistically significant. Finally, in GOG protocol 204, which studied multiple platinum-doublets in this population, it demonstrated again no difference in QoL scores between these regimens. The QoL data were published in 2010, and there were no reported differences among the four regimens. In 2015, Penson et al. reported the QoL results of GOG protocol 240, which introduced bevacizumab into the standard chemotherapy treatment of advanced or recurrent cervical cancer. Using the Functional Assessment of Cancer Therapy—Cervix Trial Outcome Index (FACT-Cx TOI), there were no significant differences in QoL. Thus, despite improved PFS and OS, this did not necessarily translate into improved QoL. The FACT-Cx TOI in this analysis was associated with survival in that for every increment of 10 on the QoL score, hazard of progression or death was also incrementally affected. This same concept of QoL scores at baseline predicting outcomes in QoL analyses has been demonstrated in multiple investigations. In fact, Chase et al. did demonstrate that on several GOG cervical cancer trials, QoL measurement alone at baseline could predict the development of toxicity, including leukopenia, anemia, and GI toxicity. This association remained significant despite controlling for important patient characteristics such as performance status, age, and treatment type. In a randomized Japanese GOG study of carboplatin and paclitaxel versus cisplatin and paclitaxel in advanced or recurrent cervical cancer, the investigators used hospitalizations as an objective measurement of QoL. The carboplatin arm demonstrated less hospitalization time, which the authors suggest might equal improved QoL of that regimen. Finally, QoL was not worsened with the incorporation of bevacizumab into traditional chemotherapy for advanced and/or metastatic cervical cancer.

Several other areas within the trajectory of cervical cancer have been explored in relation to QoL. These include preinvasive disease, surgery, radiation therapy, and survivorship. In 2013, a multicenter observational and cross-sectional study of 842 women in the United Kingdom was published. This study included women without dysplasia and genital warts as well as a spectrum of preinvasive cervical and vulvar disease. QoL assessment indicated a significant negative impact of these diseases on psychosocial well-being, including sexual functioning. Of note, anxiety and depression as well as pain were particularly noteworthy in women with genital warts. These effects on QoL not only persist throughout treatment but are also demonstrated in survivorship. For example, several studies have indicated that long-term effects can include worsening lymphedema and bowel, genitourinary, and sexual function after completion of multimodality treatment including chemotherapy or radiation. In addition, psychosocial functioning is affected in the long term. This prompted researchers to evaluate a program of telephone counseling for cervical cancer survivors. A total of 204 patients were randomized to telephone counseling versus usual care. This study not only demonstrated improvements in depression, gynecologic, and cancer-specific concerns in the intervention group but also demonstrated positive effects on biomarkers for psychosocial stress, with decreased levels of cytokines such as various interleukins. The hope is that research will continue to demonstrate improvements in PROs that can be objectively measured by biomarkers of stress and that ultimately this can link to clinical outcomes.

In the cervical cancer population, unique experiences and patient characteristics have the ability to affect outcomes. In this disease site, it is perhaps especially critical to try to alter environmental factors. Several studies investigate this. For example, increasing social support during a study of telephone counseling was associated with QoL improvements. Other research demonstrates that smoking, alcohol use, and poor physical activity is especially problematic in a population of cervical cancer patients. This is a potential intervention target as these behaviors are associated with worse QoL and patients on this trial were not meeting national guidelines in these areas. Of note, much of the toxicity of cervical cancer therapy is associated with GI toxicity. In the realm of environmental factors associated with toxicity and QoL, studies have explored relation of the microbiome to toxicity and thus QoL. Various characteristics of the gut microbiome have been described and some indicate that composition of the microbiome not diversity may be associated with treatment-related toxicities.

Quality of life in endometrial cancer

Endometrial cancer usually presents with vaginal bleeding and at an early stage. Treatment is surgical and occasionally involves postoperative chemotherapy, radiation, or both. Likely because of the less invasive or aggressive treatment, with the majority of patients presenting at an early stage, there are not as many studies that focused exclusively on QoL in patients with endometrial cancer. A Cochrane review published in 2012 but edited in 2015 included trials in advanced or recurrent endometrial cancer. Of the 14 randomized trials reported, no QoL results were identified and thus analyzed . In fact, in the GOG, there are only two published reports of QoL in patients with endometrial cancer, and there has yet to be an endometrial cancer–specific subscale incorporated into this research. In 2007, Bruner et al. published the results of the QoL components of GOG protocol 122, which compared whole-abdominal irradiation (WAI) and combination chemotherapy in advanced patients with endometrial cancer (see Table 19.5 ). In this study using the FACT-G in addition to the Fatigue Scale (FS), Assessment of Peripheral Neuropathy (APN), and Functional Alterations due to Changes in Elimination (FACE), QoL was measured before and after therapy. The WAI arm suffered greater deficits in the FS and FACE scores, but these scores somewhat improved over the varying time points. Conversely, the neuropathy scores were worse in the chemotherapy arm. Interestingly, the FACT-G scores did not differ between the two arms at any time point. The data generated in this study provide patients with the risks and benefits of each therapy and can perhaps shed light on whether QoL concerns in patients can overshadow the potential survival benefits of one modality.

In 2009, the QoL data were published from Europe’s Post Operative Radiation Therapy in Endometrial Cancer (PORTEC-2) trial, in which patients received external-beam versus vaginal brachytherapy. This study used the QLQ-C30 measure in addition to subscales from the prostate cancer and ovarian cancer module, which included both bowel and bladder symptoms and sexual functioning symptoms from those scales. Despite the lack of significant differences between groups, there was a low level of sexual functioning in the two groups. The vaginal brachytherapy group overall reported better social functioning, fewer bowel symptoms, and less limitation in activity because of these side effects compared with external-beam radiation therapy. These results suggest the use of a treatment modality potentially as effective with less impact on QoL. In another GOG trial, Kornblith et al. published the results of the QoL component of the LAP-2 protocol, which compared patients undergoing laparoscopic staging versus laparotomy. The measures used included the FACT-G as well as a six-item scale consisting of items related to surgical side effects. In addition, they used the Physical Functioning Subscale of the Medical Outcome Study-Short Form (MOS-SF36. PF), which was developed to assess activities of daily living. The differences in QoL between these two groups favored the laparoscopy group, but this difference was modest. The authors very eloquently describe that to spare 1 patient the short-term QoL decline with laparotomy, 10 patients would have to be offered laparoscopy. In fact, only better body image persisted in the laparoscopy group at 6 months; all other measurements became comparable by that time point. A more recent trial explored PROs in patients surgically staged with robotic versus laparotomy approaches. Overall QoL scores did not differ significantly between these groups at 12 months. QoL data can be complex in analysis and interpretation but can shed light on misconceptions regarding the benefits of one therapy or intervention over another.

More recently, several endometrial trials have focused on diet and exercise interventions focused on the prevalence of obesity in these patients. A systematic review and meta-analysis of body mass index and QoL in endometrial cancer included four studies with a total of 1362 patients. With the high incidence of obesity in endometrial cancer patients, this is an optimal patient population to study diet and exercise interventions. Indeed, obese endometrial cancer survivors demonstrate poorer physical, social, and role functioning compared with nonobese patients. One hundred patients with posttherapy endometrial cancer stage I to IIIA participated in a 6-month study of a home-based exercise intervention. In this study, the nonobese women demonstrated higher baseline QoL. Although the obese women had fewer improvements in QoL, both the obese and nonobese groups demonstrated some improvements in QoL with this exercise intervention over the period of 6 months. Though other research involving physical activity interventions demonstrated improvements in physical and general health, these changes were not associated with improvements in emotional or mental health. More in depth exploration into the complex interplay between physical health and emotional/social well-being is necessary in this population.

The SUCCEED trial looked at an intervention of self-efficacy, weight loss, and QoL in obese endometrial cancer survivors. This study included 75 women randomized to usual care versus an intervention. The intervention group demonstrated significant improvements in the Weight Efficacy Life-style questionnaire. Essentially, this questionnaire measures a patient’s ability to resist weight gain and maintain healthy lifestyle behaviors. However, the FACT-G differences (QoL measurement tool) between the groups were reported as minimal except perhaps in the fatigue domain.

The literature continues to explore the changes in QoL with radiation (with or without chemotherapy) in endometrial cancer patients. The Patient-Reported Outcome-CTCAE (Common Toxicity Criteria for Adverse Events) tools can help to specifically measure patient-reported treatment toxicities in some of these trials. It is clear that intensity modulated radiation therapy (IMRT) demonstrates lessened patient-reported GI and urinary toxicities. In trials where chemotherapy is added to the radiation schedule, it is now understood that both patient-reported toxicities such as neuropathy as well as more general QoL scores are worsened, which further places into question how much actual benefit the additional chemotherapy would bring to these patients. Measuring QoL in this setting demonstrates how helpful PROs can be in a setting where outcome efficacy is questionable.

Fatigue

Fatigue is perhaps the most common symptom experienced by patients with gynecologic cancers. Fatigue is the most prevalent (60%–96%) and one of the least understood symptoms that affect cancer patients. In a study of 102 ovarian cancer patients with a life expectancy of less than 6 months, fatigue was the most common disease-related symptom, followed by worry and trouble sleeping. Several factors are implicated in the causes of fatigue, for example, anemia, pain, nutrition, insomnia, cognitive dysfunction, and psychological distress. Fatigue has been reported by gynecologic patients as severe, distressing, and uncontrollable. Unfortunately, cancer-related fatigue is not relieved by rest and might not improve posttherapy. Patients have identified fatigue with cancer as the major obstacle to normal functioning and good QoL. Although almost a universal symptom of patients undergoing primary antineoplastic therapy or treatment with biologic response modifiers, it is also extremely common in populations with persistent or advanced cancer. Perhaps this is the result of inadequate attention on the part of health care professionals coupled with patients’ reluctance to discuss their fatigue.

Given the prevalence and impact of cancer-related fatigue, there have been remarkably few studies of the phenomenon. Its epidemiology has been poorly defined, and the variety of clinical presentations remains anecdotal. Perhaps this is secondary to the inability of fatigue to be measured in any way other than subjectively. The existence of discrete fatigue syndromes linked with predisposing factors or potential etiologies has not been confirmed, and clinical trials to evaluate putative therapies for specific types of cancer-related fatigue are almost entirely lacking.

Patients and practitioners can generally differentiate the “normal” fatigue experienced by the general population from the clinical fatigue associated with cancer or its treatment. The term “asthenia” has been used to describe fatigue in oncology patients but has no specific meaning apart from the more common term. This condition is inherently subjective and multidimensional. Typically, it develops over time and is characterized by diminishing energy, mental capacity, and the psychological condition of cancer patients ( Table 19.6 ). It is also linked with lethargy and malaise in the revised National Cancer Institute Common Toxicity Criteria. These classifications may enhance awareness of fatigue and improve reporting of the condition.

When fatigue is primarily related to a treatment, there is generally a clear temporal relationship between the condition and the intervention. In patients receiving cytotoxic chemotherapy, for example, it often peaks within a few days and declines into the next treatment cycle. During the course of fractionated radiotherapy, it is often cumulative and may peak over a period of weeks. Occasionally, it persists for a prolonged period beyond the end of chemotherapy or radiation treatment. The relationship between fatigue and demographic characteristics, physiologic factors, and psychosocial factors is not well defined. The specific mechanisms that precipitate or sustain the syndrome are unknown. Fatigue may present a final common pathway to which many predisposing or etiologic factors contribute ( Table 19.7 ). The pathophysiology in any individual may be multifactorial. Proposed mechanisms include abnormalities in energy metabolism related to increased nutritional requirements (e.g., caused by tumor growth, infection, fever, or surgery), decreased availability of metabolic substrate (e.g., caused by anemia, hypoxemia, or poor nutrition), or the abnormal production of substances that impair metabolism or normal function of muscles (e.g., cytokines or antibodies). Other proposed mechanisms link fatigue to the pathophysiology of sleep disorders and major depression. Further research is necessary to determine mediating mechanisms and optimal interventions.

A detailed characterization of fatigue combined with an understanding of the most likely etiologic factors is necessary to develop a therapeutic strategy ( Fig. 19.1 ). The initial approach to fatigue should involve a screening tool. The recommended screen for cancer-related fatigue begins with a simple question: How would you rate your fatigue on a scale of 0 to 10 over the past 7 days? The score generated from this scale can drive the approach to treatment. A score of 0 to 3 is considered mild, 4 to 6 moderate, and greater than 6 severe. Those in the moderate to severe range should be evaluated immediately, but mild fatigue may be reevaluated. Evaluation of fatigue can then be approached differently depending on the patient’s being in active treatment, remission, or the end of life. Patients screening positive for moderate to severe fatigue should undergo a comprehensive assessment including the description of fatigue-related phenomena, a physical examination, and a review of laboratory and imaging studies that may allow a possible hypothesis concerning pathogenesis, which, in turn, may suggest appropriate treatment strategies. Patients may describe fatigue in terms of decreased vitality or lack of energy, muscular weakness, dysphoric mood, insomnia, impaired cognitive functioning, or some combination of these disturbances. Although this variability suggests the existence of fatigue subtypes, this has not yet been confirmed. Regardless, the patient’s history should clarify the spectrum of complaints and attempt to characterize features associated with each component. This information may suggest specific causes (e.g., depression) and influence the choice of therapy. Neurologic and psychological evaluation may also help further clarify potential causes of fatigue in some patients. Other characteristics are similarly important. Onset and duration, for example, distinguish acute and chronic fatigue. Acute fatigue of recent onset is anticipated to end in the near future. Chronic fatigue is persistent for a prolonged period (weeks to months or longer), and it is not expected to remit in a short time. Patients perceived to have chronic fatigue typically require more intensive evaluation as well as a management approach focused on both short- and long-term goals. Other important descriptors of fatigue include the severity, daily pattern, course over time, exacerbating and palliative factors, and associated distress. An assessment of cancer-related fatigue should also include consideration of broader concerns, including global QoL, other symptoms, and disease status. Fatigue may be only one of numerous factors that influence QoL. Among these factors are progressive physical decline, psychological disorders, social isolation, financial concerns, and spiritual distress. Optimal care of the cancer patient includes a broader assessment of these factors and should be directed toward maintaining or enhancing QoL. Successful strategies should ameliorate fatigue within a broader approach of patient care. Evaluation of the patient regarding the nature of fatigue, options for therapy, and anticipated outcomes is an essential aspect of the therapy. Unfortunately, results of a patient survey indicate that patients and their oncologists seldom discuss fatigue.

An initial approach to cancer-related fatigue includes efforts to correct potential etiologies, if possible and appropriate. This may include elimination of nonessential centrally acting drugs, treatment of a sleep disorder, reversal of anemia or metabolic abnormalities, or management of major depression. Referring to the National Comprehensive Cancer Network (NCCN) guidelines for the approach to pain, emotional distress, anemia, and nutrition may prove helpful. Perhaps sleep disturbances, interventions aimed at increasing activity levels, and the management of comorbidities may be more straightforward. Many of these initial interventions are relatively simple and pose minimal burdens to the patient, health care provider, and caregiver.

In general, education and counselling are advised in the initial treatment of cancer-related fatigue. A meta-analysis exploring cancer-related fatigue interventions demonstrated that psychological and exercise interventions were more effective in reducing fatigue than pharmaceutical interventions. Patients and family may be requested to document fatigue levels in a diary or on a scale and to describe associations or exacerbations and timing of the symptom. Similar to the approach to urinary incontinence, patients may be counseled toward behavior modifications or adjustments to help avoid this symptom such as eliminating nonessential potentially “tiring” behaviors that may call for unnecessary expenditure of energy. Conversely, increasing physical activity has been proposed as a treatment for cancer-related fatigue. A Cochrane analysis demonstrated improvements in fatigue with an exercise intervention. Adherence to the US Surgeon General’s recommendations for 30 minutes of exercise on most days of the week may thus be advisable for cancer patients in recovery. Physical therapy referrals should be made in certain patients, and caution with activity should be obvious in certain patients such as those with bone metastasis, neutropenia, anemia, thrombocytopenia, or active infection. Therapies such as massage, acupuncture, or psychosocial interventions are also described in the NCCN guidelines. A recent meta-analysis did demonstrate that yoga practice was associated with small to moderate improvements in fatigue and depression but did not seem to impact QoL. A 6-week intervention of Swedish massage therapy (SMT) in a randomized clinical trial conducted in breast cancer patients demonstrated a significant reduction in fatigue.

In patients with fatigue-associated major depression, treatment with an antidepressant is strongly indicated. As many as 25% of cancer patients develop major depression at some point during their illnesses. Patients at greatest risk are those with advanced disease, uncontrolled physical symptoms (e.g., pain), or a previous history of a psychiatric disorder. Although the relationship between depression and fatigue is not well understood, they often occur together and both adversely affect QoL. Despite the high prevalence in the cancer population, depression is often underdiagnosed and consequently undertreated. A trial with an antidepressant is usually warranted in a patient with fatigue associated with any significant degree of depressed mood and similarly can be therapeutic when concurrent anxiety or pain exists. In addition, brief, focused psychological counseling can be helpful for several reasons when a mood disorder (e.g., depression or anxiety) and a physical symptom (e.g., pain or fatigue) co-occur. First, counseling offers the patient an opportunity to identify and express her fears, which are often driving the depressed or anxious mood. Second, the depressed or anxious mood can exacerbate existing physical symptoms such as pain. Therefore, provision of counseling has the dual benefit of reducing the mood disorder, which by extension reduces fatigue and pain. Third, brief, focused counseling can offer the patient important behavioral modifications such as time and energy management strategies that permit and teach the patient to conserve energy physically and emotionally for the priorities in their life. This is useful to address the practical challenges associated with fatigue and pain management.

Anemia may be a major factor in the development of cancer-related fatigue. Anecdotally, transfusion therapy for severe anemia has often been associated with substantial improvement in fatigue. Compared to other cancers, gynecologic cancers are especially known for being associated with severe anemia, with hemoglobin levels less than 9.9 g/dL in many cases. Not only does anemia have a significant correlation with poor performance status, but it may also be implicated in tumor sensitivity to radiation and chemotherapy. It has been suggested that an optimal oxygen level for tumors to respond to therapy is reflected by a hemoglobin level between 12 and 14 g/dL. However, recent data of increased risk of thrombotic event associated with hemoglobin level higher than 12 g/dL should be taken into consideration.

New data demonstrate the association between chemotherapy-induced mild to moderate anemia and both fatigue and QoL impairment. For example, combined data from 413 patients and three randomized placebo-controlled trials of epoetin alfa, the recombinant form of human erythropoietin (EPO), reveal that treated patients experienced a significant increase in hematocrit, a reduced need for transfusion, and a significant improvement in overall QoL. Patients with an increase in hematocrit of greater than 6% also demonstrated significant improvement in energy level and daily activities. Additional studies in patients treated with chemotherapy and radiation therapy for various gynecologic tumors confirm that epoetin alfa has positive effects on hemoglobin levels. Two large, prospective, randomized, multicenter community trials have demonstrated that patients experience significant improvement in energy levels, activity level, functional status, and overall QoL when epoetin alfa is administered as an adjunct to cytotoxic chemotherapy.

Blood transfusion is an option for the treatment of cancer-related anemia; however, transfusions are associated with poor outcomes and QoL caused by infections, allergies, or dependence on medical centers. For those reasons, many practitioners rely on growth factors, including epoetin alfa (rHuEPO—Epogen or Procrit) or darbepoetin alfa (Aranesp), to treat anemia. Kurz et al. concluded that rHuEPO increases hemoglobin levels and decreases transfusions in patients with gynecologic malignancies undergoing polychemotherapy without compromising QoL. It has even been suggested that rHuEPO may be given prophylactically in older patients (>65 years) and those with baseline hemoglobin levels of less than 10.5 g/dL who are going to be given chemotherapy consisting of carboplatin and paclitaxel. Although Aranesp has a longer half-life and therefore may be given less frequently, there are few clear advantages compared with Procrit. Unfortunately, this therapy is not without risks, and there are concerns that the risk of venous thromboembolism (VTE) is increased for these patients. Results from meta-analyses from 2008 to 2013 indicated an association between increased risk of thrombotic events and erythropoiesis-stimulating agents (ESA) usage with statistically significant risk and odds ratios ranging from 1.48 to 1.69. The increased risk for thromboembolism in patients with cancer became a black box warning in the updated Food and Drug Administration (FDA) labels and subsequently as part of the Risk Evaluation and Mitigation Strategy (REMS). The increased FDA restrictions stemmed from eight randomized studies showing a decrease in OS or decreased locoregional disease control with EPO usage, including a cervical cancer study. More recently, research has shown that survival may be decreased in women receiving rHuEPO for correction of anemia for hemoglobin levels over 12 g/dL. Therefore, currently, patients with cancer may be counseled about the risks and benefits of the use of growth factors for the correction of anemia.

The NCCN and ASCO guidelines recommends that EPO should not be recommended to patients who are receiving myelosuppressive chemotherapy with curative intent. Patients who are undergoing palliative treatment may receive EPO under REMS guidelines and with the informed consent of the patient. Many of the pharmacologic therapies for fatigue associated with medical illness have not been rigorously evaluated in controlled trials. Nonetheless, there is evidence to support the use of several drug classes. Psychostimulants, such as methylphenidate and dextroamphetamine, have been well studied for the treatment of opioid-related somnolence and cognitive impairment and depression in elderly and medically ill patients. There are no controlled studies of these drugs for cancer-related fatigue, but empiric administration may yield favorable results in some patients. The national guidelines, however, are cautious in their suggestion of these interventions.

A clinical response to one drug does not necessarily predict a response to the others, and sequential trials may be needed to identify the most beneficial therapy. Methylphenidate has been more extensively evaluated in the cancer population than other stimulant drugs and is often the first drug to be administered. It is available in a chewable formulation that can be absorbed through the buccal mucosa for patients who are unable to swallow or take oral medications.

Adverse effects associated with the psychostimulants include anorexia, insomnia, tremulousness, anxiety, delirium, and tachycardia. To ensure safety, slow and careful dose escalation should be undertaken to minimize potential adverse effects. A regimen of methylphenidate, for example, usually begins with a dose of 5 to 10 mg once or twice daily (morning and, if needed, midday). If the drug is tolerated, the dose is increased. Most patients appear to require less than 60 mg/day, but some require much higher doses.

Extensive anecdotal observations and very limited data from controlled trials support the use of low-dose corticosteroid therapy in fatigued patients with advanced disease and multiple symptoms. Dexamethasone and prednisone are most commonly used. There have been no comparative trials.

The selective serotonin reuptake inhibitors, secondary amine tricyclics (e.g., nortriptyline and desipramine), and bupropion are sometimes associated with the experience of increased energy that appears disproportionate to any change in mood. For this reason, these agents have also been tried empirically in nondepressed patients with fatigue. Given the limited experience in the use of these drugs for this indication, an empirical trial should be considered only in severe and refractory cases.

Amantadine has been used to treat fatigue in patients with multiple sclerosis, but it has not been studied in other patient populations. This drug is usually well tolerated, and an empirical trial may be warranted in selected patients with severe refractory cancer-related fatigue.

Nonpharmacologic approaches for the management of cancer-related fatigue are supported mainly by favorable anecdotal experience ( Table 19.8 ). Patient preferences should be considered in the selection of one or more of these approaches. In particular, sleep hygiene principles should be tailored to the individual patients and might include the establishment of a specific bedtime, awake time, and routine procedures before sleep. Patients should also be instructed to avoid stimulants and central nervous system (CNS) depressants before going to sleep. Whereas regular exercise performed at least 6 hours before bedtime may improve sleep, napping in the late afternoon or evening may worsen it.

Cancer and its treatment can also interfere with dietary intake. With aggressive approaches to management, the patient’s weight, hydration status, and electrolyte balance should be monitored and maintained to every extent possible. Regular exercise may improve appetite and increase nutritional appetite. Referral to a dietitian for nutritional guidance and suggestions for nutritional supplements may be useful.

The NCCN cancer-related fatigue guidelines were recently updated. This update includes further emphasizing the incorporation of fatigue as a cancer patient’s “vital sign.” Specific attention should now be paid toward improving sleep hygiene and enhancing the education of family members on cancer-related fatigue. In summary, the strategy aimed at combating fatigue should include physical activity, psychosocial interventions (i.e., mindfulness-based therapy), nutrition counselling, and sleep interventions. Pharmacologic interventions such as psychostimulants can be used, but particular attention toward medications to address comorbidities, sleep, pain, anemia, and mood is considered important.

Pain

Cancer pain can be managed effectively through relatively simple means in up to 90% of the 8 million Americans who have cancer or a history of cancer. Unfortunately, pain associated with cancer is often undertreated. Although cancer pain or associated symptoms cannot always be eliminated, proper use of available therapies can effectively relieve pain for most patients. Management of pain extends beyond pain relief and encompasses the patient’s QoL and the ability to work productively, to enjoy recreation, and to function normally in the family and society.

State and local laws often restrict the medical use of opioids to relieve cancer pain, and third-party payers may not reimburse for noninvasive pain control treatments. Thus, clinicians should work with regulators, state cancer pain initiatives, and other groups to eliminate these health care system barriers to effective pain management ( Table 19.9 ).

Flexibility is the key to management of cancer pain. Thorough discussions with the patient and their families encouraging them to be active in pain management are critical ( Table 19.10 ). Patients often need reassurance to report pain because effective treatment strategies exist. Failure to assess pain is a critical factor leading to undertreatment. Thus, a simple screening tool should be used to inquire about the patient’s pain level. This is described as “universal screening” by the NCCN. The goal of the initial assessment of pain is to characterize the pain by location, intensity, and etiology. This can be accomplished through a detailed history, physical examination, social assessment, and diagnostic evaluation. The mainstay of pain assessment is patient self-reporting. To enhance pain management across all settings, clinicians should teach patients to use pain assessment tools in their homes. Clinicians should listen to the patient’s descriptive words about the quality of the pain, inquiring about its location, severity, and aggravating or relieving factors and the patient’s cognitive response to the discomfort. Finally, goals for pain control should be clear with active engagement of the patient in the decision process. Continued assessment of cancer pain is crucial. Changes in pain patterns and the development of new pain should trigger diagnostic evaluation and modification of the treatment plan. Persistent pain indicates the need to consider other etiologies (e.g., related to disease progression or treatment, and alternative—perhaps more invasive—treatment; Fig. 19.2 ). Pain related to an oncologic emergency should be addressed immediately. This includes pain resulting from a bone fracture or impending fracture, brain or epidural metastases, infection, or obstructed or perforated viscous.

Drug therapy is the cornerstone of cancer pain management. It is effective, relatively low risk, and inexpensive and usually works quickly. Even within the same family of analgesic drugs, individual variations in tolerability and side effects are well recognized. Recommendations for pharmacologic therapy begin with the World Health Organization (WHO) ladder ( Fig. 19.3 ), a three-step hierarchy for analgesic pain management. Substitution of drugs within a category should be tried before switching therapy. The simplest dosage and schedule as well as the least invasive pain management modality should be attempted first. For mild to moderate pain, nonsteroidal antiinflammatory drugs (NSAIDs) (WHO ladder step 1) are often effective (see Table 19.10 ). When pain persists or increases ( Table 19.11 ), opioids can be added (WHO ladder step 2). Moderate to severe pain requires opioids of higher potency and dose (WHO ladder step 3) ( Table 19.12 ). Dosing should be on a regular schedule (i.e., “by the clock”) to maintain a level of drugs that would help prevent the recurrence of pain. Ask for patient and family cooperation in establishing the effective level when administering medications to prevent long-term cancer pain on an around-the-clock basis with additional doses (“as needed” and usually required).

Oral administration is preferred because it is convenient and usually cost effective. When patients cannot take oral medications, other less invasive (e.g., rectal or transdermal) routes should be offered. Parenteral methods should be used only when simpler, less demanding, less costly methods are inappropriate or ineffective. An assessment of the patient’s response to several different oral opioids is usually advisable before abandoning the oral route in favor of parenteral, neurosurgical, or other invasive approaches. Rectal administration is a safe, inexpensive, and effective route for the delivery of opioids as well as nonopioids when patients have nausea or vomiting. Rectal administration is inappropriate for patients who have diarrhea, anal or rectal involvement, or mucositis; are neutropenic or thrombocytopenic; are physically unable to place the suppository in the rectum; or who prefer other routes. Transdermal administration is also feasible but does not allow rapid dose titration. Patient-controlled analgesia (PCA) devices can be used both on an inpatient or outpatient basis. The opioid may be administered orally or via a dedicated portable pump to deliver the drug intravenously, subcutaneously, or epidurally (intraspinally). Intraspinal administration should be considered for patients who develop intractable pain or intolerable side effects from other routes of administration. Use of this route requires skill and expertise that may not be available in certain settings. Table 19.13 presents the advantages and disadvantages of regional administration. This route is often efficacious because gynecologic tumors often affect the pelvis, making profound analgesia frequently possible without motor or sympathetic blockade. Drugs and routes of administration that are not recommended for the management of cancer pain are summarized in Table 19.14 .

The general principles to pain management are described by the NCCN and were updated in 2015. This includes a dose of medication that sustains pain relief for the full time period before the next dose. The dose should be calculated based on the patient’s requirements for the previous 24 hours. A sustained dose with an around-the-clock medication should be started in addition to an as needed medication. The around the clock and as needed doses are increased based on the severity of the uncontrolled pain. Of note, noncombination opioid preparations are preferred to eliminate the risk of overdosing acetaminophen or ibuprofen. The “as needed” medication should be about 10% to 20% of the sustained total 24-hour dose. Before converting to a transdermal approach, the pain should be relatively well-controlled by an around-the-clock opioid pain regimen. The maximum fentanyl transdermal dose is 100 mcg/h, but if a patient requires a higher dose, multiple patches may be used. An IV dose of 0.67 mg/day of IV morphine or 2 mg/day of oral morphine is equivalent to 1 mcg/h of transdermal fentanyl. Although the patch lasts typically for 72 hours, cachectic patients, fever, or heat from external sources such as electric blankets may increase the absorption of the transdermal fentanyl and thus may be contraindications to patch use. Some patients may require a new patch every 48 hours. Oxycodone or morphine formulations continue to demonstrate equivalency in controlling cancer-related pain with perhaps less hallucinations in the patients treated with oxycodone versus morphine. The idea of opioid rotating to decrease cost, intolerance, or a change in the patient’s condition is suggested.

Clinicians who follow patients during long-term opioid treatment should watch for potential side effects and administer agents to manage them, and, in some cases, prophylactically. Constipation, as well as nausea and vomiting, both common side effects to opioid analgesics, are discussed later. Drug-induced sedation is common and can be treated by a reduction in opioid dose or use of low-dose CNS stimulants such as caffeine, methylphenidate, dextroamphetamine, or modafinil in the morning or early afternoon may be considered. Patients should also be assured that tolerance may develop over time, especially after an opioid dose escalation. Patients receiving long-term opioid therapy generally develop tolerance to the respiratory depressant effects of these agents. When indicated for reversal of opioid-induced respiratory depression, administration of naloxone is recommended with titration in small increments to improve respiratory function without reversing analgesia. Careful monitoring is mandatory until the episode of respiratory depression resolves. For more subacute respiratory depression, simply withholding one or two doses until the symptoms resolve followed by restarting at 25% of the total dose is often effective. A combination of opioids and NSAIDs is frequently used to optimize cancer pain management, especially in patients with bone and lymphatic involvements. Dry mouth, urinary retention, pruritus, myoclonus, altered cognitive function, dysphoria, euphoria, sleep disturbances, sexual dysfunction, physiologic dependence, tolerance, and inappropriate secretion of antidiuretic hormone are also reported side effects of opioid agents.

Adjuvant drugs are valuable during all phases of pain management to enhance the analgesic efficacy, treat concurrent symptoms, and provide independent analgesia for specific types of pain. These adjuvants include corticosteroids, anticonvulsants, antidepressants, neuroleptics, local analgesics, hydroxyzine, and psychostimulants. Corticosteroids provide a range of effects, including mood elevation, antiinflammatory activity, antiemetic activity, and appetite stimulation, and may be beneficial in the management of cachexia and anorexia. They also reduce cerebral and spinal cord edema and are essential in the emergency management of elevated intracranial pressure and epidural spinal cord compression. Anticonvulsant agents are used to manage neuropathic pain, especially lancinating or burning pain. They should be used with caution when administered to patients undergoing marrow suppressant therapy such as chemotherapy and radiation. Tricyclic antidepressants (TCAs) are useful in the pharmacologic management of neuropathic pain. These drugs have innate analgesic properties and may potentiate the analgesic effects of opioids. Perhaps the most widely reported experience has been with amitriptyline, a tertiary amine, which may be more efficacious but secondary amines (i.e., nortriptyline, imipramine) are usually better tolerated. The use of serotonin–norepinephrine reuptake inhibitors such as duloxetine and venlafaxine have increased because of better tolerance. More frequently, anticonvulsants have become more popular, especially with the availability of pregabalin, which has a more desirable side effect profile than the antidepressants and less dizziness than its parent drug, gabapentin. Local analgesics have been used to treat neuropathic pain but are limited in their effectiveness without concurrent use of an opioid, antidepressant, and/or anticonvulsant. Psychostimulants, as discussed earlier, may be useful in reducing opioid-induced sedation when opioid dose adjustment (e.g., reduced dose and lengthening dosing intervals) is not effective. Unfortunately, despite several randomized trials having been completed, cannabinoids have not demonstrated efficacy compared to placebo in the control of cancer-related pain. Furthermore, cannabinoids are associated with more adverse events such as somnolence and dizziness.

Patients should be encouraged to remain active and participate in self-care when possible. Noninvasive physical and psychosocial modalities can be used concurrently with drugs and other interventions to manage pain during all phases of treatment. The effectiveness of these modalities depends on the patient’s participation and communication concerning which methods best alleviate pain. Generalized weakness, deconditioning, and aches and pains associated with cancer diagnosis and therapy may be treated by cutaneous stimulation such as heat or cold, massage, pressure, and vibration. Unfortunately, these modalities sometimes increase pain before relief occurs. Massage should not be substituted for exercise in ambulatory patients. Exercise is useful for treating subacute and chronic pain because it strengthens weak muscles, mobilizes stiff joints, and helps to restore coordination and balance, thus enhancing patient comfort and providing cardiovascular conditioning. Physical therapists may be consulted to increase weightbearing exercise. Repositioning is also effective to maintain correct body alignment and prevent or alleviate pain and possibly prevent ulcers. Immobilization is only effective to stabilize fractures or otherwise compromised limbs or joints. Finally, acupuncture, which involves inserting small solid needles into the skin, may be an effective alternative to more standard therapies. There have been many randomized trials exploring acupuncture/acupressure and results have been consistently in favor of these modalities to both reduce pain and the use of traditional analgesics including opioids.

Cognitive behavioral interventions are an important part of a multimodal approach to pain management. They help to give the patient a sense of control and to develop appropriate skills to deal with pain. These skills include relaxation and imagery, cognitive distraction and reframing, patient education, psychotherapy, biofeedback, structured support, and support groups and pastoral counseling.

With rare exceptions, a less invasive analgesic approach should precede invasive palliative approaches. However, for a few patients in whom behavioral, physical, and drug therapy do not alleviate pain, invasive therapies are useful. These include radiation therapy to destructive bone metastasis, palliative surgical approaches, and nerve blocks.

Because vulvar cancers and occasionally ovarian or endometrial cancers occur in elderly patients, they especially require comprehensive assessment and aggressive management when cancer pain occurs. Older patients are at risk for undertreatment of pain because of underestimation of their sensitivity to pain, the expectation that they tolerate pain well, underreporting because of declined cognitive functions, and the misconceptions about their ability to benefit from the use of opioids. Careful consideration should be given to elderly patients who are in pain and who have multiple chronic diseases that increase their risk for drug–drug and drug–disease interactions. In addition, visual, hearing, motor, and cognitive impairments may require simpler pain assessment scales and more frequent pain assessments. NSAIDs are more likely to cause gastric and renal toxicity and other drug reactions such as cognitive impairment, constipation, and headaches in older patients. Alternative NSAIDs (e.g., choline magnesium trisalicylate) or co-administration of misoprostol should be considered to reduce gastric toxicity. Older persons tend to be more sensitive to the analgesic effects of opioids. In addition, the peak opioid effect is higher, and the duration of pain relief is longer. Drug clearance may also be slower, thus making cautious initial dosing and subsequent titration and monitoring necessary. Elderly patients may not be able to physically place rectal suppositories or activate PCA devices. Because of the addictive property of opioid analgesics, the NCCN and the FDA have incorporated a detailed algorithm, the REMS, for monitoring patients for both safety and abuse. Transdermal, transmucosal, and long-acting opioids are included in this list of highly monitored medications, and health care providers are encouraged to identify patients at risk for abuse or misuse.

Nausea and vomiting

Nausea and vomiting also have a high prevalence in patients with advanced cancer. The cause may be divided into physiologic, treatment-related, metabolic, and psychological causes. As is the case in many symptoms of advanced cancer, the causes of nausea are often multifactorial. More than 50% of patients receiving opioids experience nausea in the first 10 to 14 days of therapy until they develop tolerance to this side effect. Many palliative specialists may prescribe antiemetics for the first weeks of opioid therapy. The most successful therapy is given orally and prophylactically. Nausea and vomiting tend to be the most feared symptom related to cancer treatment. Chemotherapy-induced nausea and vomiting (CINV) may occur in 70% to 80% of patients receiving chemotherapy. CINV can be divided into four subcategories. The NCCN defines acute CINV as occurring shortly after chemotherapy administration but resolving within the first 24 hours. Delayed CINV occurs after 24 hours, peaks 48 to 72 hours, and resolves by 6 to 7 days. Anticipatory CINV is described as a learned or conditioned response before chemotherapy, and finally, breakthrough CINV can occur when a patient experiences symptoms despite appropriate therapy.

An evaluation and approach to the management of patients with chemotherapy-induced emesis appears in Fig. 19.4 . Agents used to control nausea and vomiting have different mechanisms of action and may be used in combination for better control ( Table 19.15 ). There are basically three main classes of drugs for CINV: 5-HT serotonin receptor antagonists, corticosteroids, and the neurokinin-1 (NK ₁ ) antagonists. The ASCO recently updated its guidelines for the use of antiemetics in oncology; these recommendations, with some additions, are detailed in Fig. 19.5 . All have a high therapeutic index for CINV and should be considered first-line therapy. The 5-HT3 antagonist blocks serotonin receptors of the peripheral GI tract and the chemoreceptor trigger zone (CTZ) to prevent vomiting. Corticosteroids are often used to enhance the effect of other agents, yet they have an unclear antiemetic action. It has been postulated that corticosteroids overcome the adrenal insufficiency that causes nausea and vomiting in patients with advanced cancer. Dexamethasone (20 mg IV) is most effective when administered with a 5-HT3 antagonist. There are several 5-HT serotonin receptor antagonists that all have similar therapeutic and side effect profiles. Several large, multicenter, double-blind, randomized phase III trials have shown that palonosetron performs superior to other agents in this class in preventing emesis associated with both moderate and high emetic risk chemotherapy regimens, particularly for delayed emesis. The three-drug combination of 5-HT receptor antagonists in addition to dexamethasone and receptor antagonists are indicated for the control of CINV in high-dose chemotherapy regimens. NK ₁ receptor antagonist plays a significant role as an adjunct for acute CINV control and single agent for delayed CINV control. For breakthrough CINV, the following agents may be considered as a replacement or adjunct agent when first-line therapeutics fail: metoclopramide, phenothiazines, butyrophenones, and cannabinoids. Metoclopramide has its effect locally in the gut and is excellent to control vomiting because of gastric stasis. In chemotherapy prophylaxis, it can be combined safely with steroids or benzodiazepines. Phenothiazines partially inhibit the CTZ and are the drugs of choice for mild CINV. Ondansetron is the drug of choice for radiation-induced nausea and vomiting. Benzodiazepines are indicated for anticipatory vomiting and can be used in patients with nausea caused by high anxiety. The amnesic effects of benzodiazepines are helpful to many patients. Olanzapine, an atypical antipsychotic, has been found to be effective for CINV, especially for patients with high rates of nausea. Randomized trials and meta-analysis continue to demonstrate improvements in the control of CINV with olanzapine with some noteworthy side effects such as somnolence and fatigue. Olanzapine is now incorporated into the preferred treatment options outlined by the NCCN for highly emetogenic regimens. Patients who describe around-the-clock nausea, not just intermittent with meals, may have a CNS lesion or a proximal small bowel obstruction that must be evaluated. Cannabinoids are best used in younger patients and have equivalent or superior activity compared with the phenothiazine agents. Specifically, the cannabinoids dronabinol and nabilone, both approved by the FDA, are indicated for refractory CINV and despite dysphoric side effects hindering their prescription, patients have been shown to prefer the use of these agents for subsequent chemotherapy cycles. In anticipatory CINV, the most important concept is the prevention of acute and delayed emesis with the above drugs. However, ultimately when one is faced with controlling anticipatory CINV, benzodiazepines and behavioral therapy are suggested. Finally, when CINV proves to be refractory, one must consider other causes of nausea and vomiting. With the change of legislature, the use of medical marijuana may be sought by patients. Although the route of administration (inhalation, nasal, and edible) may all be effective. Physicians must become familiar with state and federal requirements before prescribing to assure patient safety and to honor professional oath. Finally, the use of ginger and more specifically ginger extract has been suggested and now studied for CINV. Ginger extract has been associated with both improvement of CINV as well as fatigue with a very limited side effect profile.

Last, an important concept in CINV is the selection of therapy based on the risk assessment of the chemotherapeutic agents used, a concept highlighted in the recent ASCO guidelines. The emetogenicity of these antineoplastic drugs is separated into four categories: high, moderate, low, and minimal risk. For example, a high-risk drug such as cisplatin requires combination treatment with 5-HT receptor blockers, NK ₁ receptor antagonist, and corticosteroids followed by corticosteroid and NK ₁ receptor antagonist for delayed CINV. Low emetogenic drugs (e.g., paclitaxel) may only require dexamethasone without additional agents for acute CINV.

Diarrhea and constipation

Diarrhea is a common complication of pelvic radiation and systemic chemotherapy. It can generally be managed with intestinal opioid receptor agonist such as loperamide. Rarely, tinctures of opium or octreotide are required for refractory or secretory diarrhea ( Table 19.16 ). Patients who have been previously treated with antibacterial therapy or are immunocompromised should be tested for Clostridium difficile (e.g., enzyme-linked immunosorbent assay [ELISA] for enterotoxin) before antidiarrheal therapy. When detected, metronidazole or vancomycin therapy is usually effective. Rarely, cholestyramine is needed to control the symptoms of C. difficile infectious diarrhea. A search for more unusual infectious etiologies such as salmonella, shigella, Escherichia coli, and parasites has not been cost effective among gynecologic oncology patients in the absence of foreign travel or other data, which might suggest these etiologic agents.

Constipation is highly prevalent in patients with advanced cancer. Close to 90% of patients receiving opioids have difficulty passing stool or pass stool infrequently. One cardinal rule of palliative care is “the hand that writes the opioids writes the laxative at the same time.” Other causes of constipation are advanced disease, other medications such as anticholinergic agents, inability to eat a high-fiber diet, insufficient fluid intake, and lack of activity.

A specific discussion regarding opioid-induced constipation is warranted here. Of note, a tolerance to the opioid used with regard to constipation does not occur. In general, a mild stimulant laxative and a stool softener should be used prophylactically, for example, senna with docusate. Adequate fluid intake and physical activity should be encouraged. Such agents as Metamucil that increase fiber content in the stool are not recommended. If constipation persists, secondary etiologies, such as bowel obstruction, should be considered. Impaction should be ruled out, and then secondary anticonstipation agents should be considered. These agents include magnesium hydroxide or citrate, bisacodyl, lactulose, sorbitol, polyethylene glycol, naloxegol, and methylnaltrexone. Various enemas or agents such as metoclopramide may also be added. A general approach should be to begin patient on a stool softener and stimulant; if in 48 hours no bowel movement results, then additional agents should be started such as Milk of Magnesia or lactulose. If no bowel movement in 72 hours, then fecal impaction should be assessed and treated. If no impaction is present, then such interventions as an enema, mineral oil, or magnesium citrate should be tried. If impacted, a glycerin suppository or oil retention enema should be used.

Metabolic causes of constipation include hypercalcemia and hypokalemia. Evaluation of complaints of constipation should include consideration of bowel obstruction and spinal cord compression. A manual rectal examination is critical when evaluating this complaint.

Several treatment options are now available for nonsurgical candidates with bowel obstructions secondary to advanced cancer. A discussion of these treatments follows.

Strategies for breaking bad news and preserving hope

Perhaps the most important tool in caring for patients and their families is effective communication. Transferring information about the diagnosis, prognosis, risks, and benefits of treatment and progression of disease is a difficult and unavoidable responsibility. Sharing bad news and responding to the questions presented by the patient and family require compassion, empathy, and skill. Unfortunately, development of these skills has not historically been emphasized in medical training even though powerful medical institutions and societies (e.g., ASCO, IOM, American Cancer Society, NCI) demand excellence and vigilance when caring for people with progressing cancer. The absence of an integrated formal palliative care curriculum throughout medical training continues to promote a skill set gap.

Identification of cancer progression, or progressive cancer-related symptoms, is often the catalyst for planning alterations in care and readdressing issues about the patient’s and family’s goals and objectives. Breaking bad news is a difficult and emotionally laden task for the physician. Thoughtful planning and delivery of such information, however, can shape the patient’s subjective experience of the physician and perception of his or her degree of support and caring. The individual’s response to news is often determined by the adequacy of information given and whether the person delivering the information showed and responded to the patient’s concerns.

Although wishes vary from one individual to another, most patients (80%) want to know their diagnosis, their chance of cure, and the adverse effects of any treatments. Significantly, patients do vary in the degree of information that they are able to assimilate at one time. Traditionally, men with advanced disease, older patients, and individuals from lower socioeconomic backgrounds are likely to want to hear fewer details and may comfortably defer to the physician or family with regard to decision making. In most cases, patients want to hear the diagnosis from the physician who will be responsible for their care.

When breaking bad news, it is critical to project reasonable hope and confidence. Despite the most sensitive and effective communication style, patients are left with major concerns about their situation as it relates to their physical, social, psychological, and spiritual well-being. Open communication should be maintained, and these concerns should be explored throughout the continuum of care using expertise and support from other team members. Excessive concern about the need to preserve hope can lead to false optimism or less than full disclosure over time. This approach leads to poor coping skills and failed expectations. Lending strength does not require that the physician be less than honest but that the truth be disclosed over a period of time in a setting where the patient has the support required and with repetition so that information can be assimilated and understood in small amounts. Patients should be encouraged and nurtured to develop their own coping strategies and defense mechanisms in dealing with their illness. This strategy brings hope against all odds. For example, if asked, “Doctor, I know I am dying, but can there still be a miracle?” The answer is, “Yes, there can be.” This suggests that the medical information is clearly understood but demonstrates confidence to overcome suffering and despair. However, it is unusual for patients to so clearly articulate their inevitable demise; more likely, they would ask, “Is there still hope?” The physician’s capacity to appropriately convey hope through articulation of a comprehensive care plan will help the patient to recognize that quality time is a reasonable expectation. Forcing a patient to adopt the physician’s view of the odds (chance of a cure) does not take into consideration what can generally harm the personal framework that patients must deal with. It also corrodes the patient–physician relationship that must endure through the palliative therapy or future care. Thus, maintaining hope is best achieved through honesty, cautious optimism, compassion, and acceptance of the vulnerability experienced by all cancer patients.

Timing is critical to breaking bad news. Judging what portion of the information should be disclosed is an art. Conveying a balance of honesty and maintaining hope can sometimes seem impossible, often because physicians tend to project their own concerns onto patients and assume that they will give up hope. In reality, hope is an innate characteristic that is rarely abandoned by open and compassionate discussion of the prognosis and treatment options. When there are no remaining viable cytotoxic treatment options, there are still many palliative options that can achieve meaningful therapeutic ends, thus preserving hope and avoiding a sense of helplessness. It is useful for patients to hear such optimism in the face of a generally poor prognosis. As the patient’s illness advances, feelings of helplessness because of a loss of control and fear of dying may resurface. At this point, the physician can openly recognize the patient’s fear and grief and externally acknowledge his or her own attitudes toward death. Unfortunately, doctors and other staff tend to spend less time with patients as the disease progresses and patients deteriorate into a terminal condition. This is presumably because the health care professionals feel helpless (and perhaps even threatened) when confronted with the imminence of death. Many health care professionals fail to realize the importance of compassion and active palliative care. Actively participating in symptom management and compassionate listening helps to ease the patient’s mind, thus allowing the patient to ask questions that help plan realistically for the future and maintain a sense of control. Although most patients know when they are dying and can verbalize this realization, they also retain hope. The physician is uniquely poised to encourage and reinforce the patient’s hope without giving false or insincere reassurance. One study that focused on end-of-life preferences of gynecologic cancer patients reported that only 5% of 108 patients would stop fighting after receiving a poor prognosis with no medically recommended treatment options left. Most (70%) of these patients expressed their resolve to continue fighting against their disease, even under the poorest prognostic circumstances. Proper management involves directing resources to assist with this fight and fostering this hope among dying patients. The involvement of other professionals familiar with care for the dying can relieve the physician of being the sole provider of hope with sensitive attention to incurable problems.

Pleural effusions

Dyspnea is the subjective feeling of respiratory distress and thus is often an interplay between anxiety and hypoxia from tumor burden. Some have suggested that supportive care of this symptom should include the provider’s ability to address the patient’s subjective breathlessness. Because dyspnea may indicate a poorer prognosis and shorter interval to death, addressing this symptom is paramount to enhancing QoL in the time period near death.

Dyspnea may result from a combination of three different complications: (1) obstructive pathology such as pleural effusions, (2) cancer cachexia or malnutrition and weakness, and (3) an increase in ventilator requirements such as metabolic acidosis or anemia. Perhaps gynecologic oncologists most frequently encounter this tumor effect in the ovarian cancer patients because the dyspnea may be secondary to pleural effusions. It is suggested that thoracocentesis and chemical pleurodesis be reserved for palliative situations in which chemotherapy or other treatments are not likely to reverse the effusions. For symptomatic control or when mechanical control of the pleural effusion is not an option, most advocate opioids, oxygen therapy, or both. Benzodiazepines as well as oxygen therapy may also have a role in control of dyspnea, although evidence is lacking. Finally, some have advocated activity modifications to prevent the onset of dyspnea, which may involve caregiver instruction to tailor activities of daily living.

Small bowel obstruction

Although epithelial ovarian cancer is a surface-spreading disease that rarely invades vital organs, partial or complete bowel obstruction is often seen at the time of initial diagnosis or, more frequently, in association with recurrent disease. Obstruction may be secondary to extrinsic compression of the small bowel or hypoperistalsis caused by mesenteric and bowel surface implants. The symptoms, which are almost always present, are intestinal colic, continuous abdominal pain, nausea, and vomiting. In most cases, nausea and vomiting can be relieved by conservative measures, and intestinal symptoms usually resolve after primary cytoreductive surgery and multiagent cisplatin-based cytotoxic chemotherapy. Despite the high objective response rate of primary epithelial ovarian cancer to several platinum-based combination chemotherapeutic regimens, most patients with advanced disease often develop IP recurrences and require a salvage regimen for palliation of symptoms. Several therapeutic strategies have been used in these patients, such as including retreatment with cisplatin or carboplatin or the use of paclitaxel, hexamethylmelamine, oral etoposide, tamoxifen, gemcitabine, liposomal doxorubicin, vinorelbine, or topotecan. Multiple clinical factors must be considered when selecting the most appropriate salvage therapy. Unfortunately, in patients previously treated with cisplatin, other therapeutic agents are not likely to be effective in relieving symptoms of bowel obstruction or ascites.

For most patients who present with bowel obstruction secondary to recurrent IP cancer, initial management should include proper radiographic documentation of the obstruction, hydration, correction of any electrolyte abnormalities, parenteral alimentation, and intestinal intubation and decompression. There is a fair representation of various treatment options in the literature. A recent retrospective study found that surgery carries a higher morbidity, although chemotherapy and surgery had similar outcomes in terms of reobstruction. Of note, conservative management alone has much earlier reobstruction rates. In some patients, the obstruction may be relieved with this conservative approach. However, palliative surgery is usually considered in almost every patient at some time before disease progression and death. When surgery is indicated, the type of surgery depends on the extent of the disease as well as on the number and location of obstructions. If the obstruction is mainly contained in one area, this area can be either resected (if secondary cytoreduction is indicated) or bypassed. Because the success of secondary cytoreduction depends on the chemosensitivity of the residual disease present after debulking surgery, intestinal bypass is generally preferable rather than resection because most patients present after multiple failed attempts at cytotoxic therapy resulting in chemotherapy-resistant cancer. In addition, intestinal bypass surgeries such as enterocolostomy are usually associated with reduced morbidity compared with a radical resection. At the time of operation, the balloon at the end of a long intestinal tube can often be palpated and used to identify the small bowel proximal to the obstruction. This small bowel can then be anastomosed in a side-to-side fashion to the most appropriate area of colon, thus bypassing the site of obstruction. Obviously, it is critical to obtain a preoperative Gastrografin enema to ensure that there is no obstruction of the lower colon beyond the bypass site. In rare cases, the colon may be encased in tumor, necessitating a colostomy with or without bypass surgery. Sadly, multiple sites of obstruction are common in patients with recurrent epithelial ovarian cancer. When multiple sites of obstruction are present, resection of several segments of intestine is usually not indicated. In inoperable cases or situations when the patient or oncologist decides to proceed with medical or conservative management, several options exist. A percutaneous endoscopic gastrostomy allows patients to hydrate themselves by mouth. By decreasing intestinal secretions, octreotide has improved QoL and lessens hospitalization. Moretti et al. used temporary nasogastric drainage with the simultaneous use of octreotide and antiemetic and analgesics in a terminal patient. Critical in their algorithm is fluid management because they stress keeping daily water intake to between 0.5 and 1.5 L. In the last month of life, this combined medical and conservative approach decreased pain, nausea, dry mouth, thirst, dyspnea, feelings of abdominal distention, and drowsiness in the palliative setting. In such cases, an ileostomy or even a proximal jejunostomy may be necessary to provide adequate intestinal diversion. On the contrary, if the extent of disease is so great that the morbidity of intestinal surgery seems excessive, stomach decompression with a gastrostomy may be effective in palliating symptoms of obstruction and ascites such as pressure, nausea, vomiting, and pain. When the stomach and anterior abdominal wall are not involved with tumor, drainage tubes can often be placed percutaneously, thus avoiding the morbidity of a laparotomy.

With careful attention to nutrition, chemosensitivity or resistance of disease, and the sites of intestinal obstruction, some degree of palliation can generally be obtained with surgery, chemotherapy, or gastric decompression. Careful attention to each of these factors is necessary to avoid unnecessary morbidity and unindicated surgery in these debilitated terminal patients with recurrent or refractory ovarian cancer.

Many have attempted to define the prognostic factors which predict benefit after surgery and some of the key factors include age older than 65 years, nutritional status, tumor burden, rapidly reaccumulating ascites, poor nutritional status, carcinomatosis, previous chemotherapy for recurrence, and radiation therapy to the whole intestine. Krebs and Gopelrud used a predictive index by scoring patients with a value of 0, 1, or 2 in six of these categories and found that patients who scored higher than six had worse outcomes with surgery. However, if surgery is chosen, the patient must understand the possibility of living with an ileostomy in roughly 50% of cases with potential complications of fistula(s) for the balance of the patient’s life. In a study by Mangili et al., only surgical treatment was a prognostic factor for improved survival; however, it is likely that patients who were offered surgery had other qualities that predisposed them to improved survival.