Chemotherapy for cancer

Assessment Objectives

This report aims to determine whether the scientific literature supports the use of chemotherapy to treat several important cancer types and to assess the weight of the evidence. Another aim is to determine, through a survey of current practices, whether the use of chemotherapy complies with the scientific evidence presented in the literature.

Assessment Strategy

Systematic literature review

Primary Data Collection

Systematic search in MEDLINE through 1998, and in Cancerlit and the Cochrane Library. Important randomized studies published through August 2000 were also assessed. Complementary information sources include the PDQ database and reference lists in published articles and abstracts.

Data Criteria: Type of Studies Included

The literature search covered 12 different tumor types included under 3 different areas of investigation:

1. Tumor types with the greatest number of patients receiving chemotherapy: breast cancer and certain malignant blood and lymphatic diseases.
2. Tumor types where the use of chemotherapy has increased during the past decade; non-small cell lung cancer, stomach cancer, pancreatic cancer, colorectal cancer, and urinary bladder cancer.
3. Ovarian cancer, in part because a new, extremely expensive group of chemotherapy drugs, taxanes, has shown very promising preliminary results.

Most studies consist of meta-analyses, randomized prospective studies, and other prospective analyses with well-defined study protocols.

Review of Publications

All chapters were extensively reviewed within the project group. Each chapter on diagnosis was also reviewed by one or more international experts within the field concerned. Four external referees, the SBU Scientific Advisory Committee, and the SBU Board of Directors have reviewed the report, which was revised according to their viewpoints and suggestions. The SBU Board of Directors then approved the summary and conclusions.

SBU Summary and Conclusions

Cancer is a general term used to describe many different types of malignant tumours. Cancers have the capacity for uncontrolled growth and can spread to different parts of the body. In Sweden, over 40 000 individuals per year (approximately as many men as women) are diagnosed with some form of cancer.

Cancer is perceived to be more frightening and threatening than many other diseases. However, the most common cause of death in both men and women in Sweden is not cancer, but diseases of the circulatory organs, accounting for more than 50% of all deaths annually. Cancers account for 23%. In people aged 45 through 64 years, the mortality rate from cancer is higher, or 40%. Cancer appears in all ages, but is most common among the elderly. One half of all patients diagnosed with cancer are above 70 years of age. Since the number of elderly is rising, increasingly more people will be diagnosed with cancer.

Cancer is usually treated by surgery, radiotherapy, and drugs - mainly agents that prevent cell growth and proliferation (cytostatics) and those that have hormonal effects - or a combination of these therapies. In recent decades, great knowledge has been gained concerning growth regulation in normal cells and cancer cells and changes in the genetic material that causes a transition to cancer. In the future, this knowledge can promote new, and hopefully better, treatment methods.

The term “cytostatics” is a general term referring to drugs that have the capacity to kill tumour cells or inhibit their growth. In Sweden, around 50 substances classified as cytostatics are currently registered by the Swedish Medical Products Agency. These drugs are usually administered directly into the bloodstream, but can also be administered in other ways.

Chemotherapy (treatment with cytostatic drugs) can lead to cure for several types of cancer and can alleviate symptoms and extend life in many patients with different forms of cancer. Cytostatic agents not only affect cancer cells but also the growth of normal cells. Hence, treatment carries a major risk for side effects. Different cytostatic drugs have different side effects, and the individual’s reaction to chemotherapy varies. A characteristic common to most cytostatics is that they affect the bone marrow and reduce the number of blood cells. This may increase the risk for severe infections.

The term “curative treatment” is used when the intent of chemotherapy is to cure the patient. The term “palliative treatment” is used when the intent is to ameliorate the patient's pain and other symptoms, but where the potential for cure does not exist. The goal of palliative treatment is to improve the patient's quality of life and extend survival. In the individual case, the different approaches to treatment may overlap.

Chemotherapy is usually delivered as a series of treatment cycles separated by an interim of one to several weeks so the body's normal cells have the opportunity to recover between cycles. Often 6 to 9 treatment cycles are administered during a 4- to 6-month treatment program, but there are many examples of both shorter and longer treatment programs. In some cases, cytostatic drugs are delivered in high doses to achieve a much greater impact on tumour cells.

Scope of This Report

This report addresses standard-dose and high-dose chemotherapy. New treatment principles currently under development, e.g., the use of drugs to prevent cancer or prevent the formation of new blood vessels and metastases (cancer spread) are not covered in this report.

The literature review has been limited to:

• Tumour groups for which chemotherapy is most commonly used in Sweden: breast cancer and some malignant blood and lymphatic diseases; acute myelogenous leukaemia, chronic lymphatic leukaemia, the most common non-Hodgkin’s lymphocytic, and Hodgkin's disease.
• Tumour types for which chemotherapy has increased in recent decades, i.e., non-small cell lung cancer, gastric cancer, pancreatic cancer, colorectal cancer, and urinary bladder cancer.
• Ovarian cancer where a new type of expensive chemotherapy, i.e., taxans, has been widely used.

The systematic literature review did not cover the side effects from chemotherapy nor the agents used to prevent or inhibit certain side effects.

Practice

A survey of the practices in chemotherapy was conducted in two healthcare regions in Sweden in 1997. The survey covered 1590 patients.

The survey of practices revealed that chemotherapy varies widely among different tumour types and covers from 10% to nearly 100% of the patients. Chemotherapy was used mainly against those types of cancer where treatment has been most effective, mainly tumours in blood-forming organs, haematological malignancies. Other treatment alternatives are often lacking in these cases. Chemotherapy was used to a lesser extent to treat other types of cancer where surgery and/or radiotherapy play a greater role.

The survey also showed cure to be the goal of treatment in approximately half of the patients who received chemotherapy. This percentage was somewhat higher at university or regional hospitals than at other hospitals. Curative treatment focused mainly on patients with aggressive malignant lymphomas, acute myelogenous leukaemia, and cancer in the breast and ovaries. Chemotherapy represents the main type of treatment for lymphoma and leukaemia. Chemotherapy was used to complement surgery and, at times, radiotherapy in treating breast cancer, ovarian cancer, and colorectal cancer. Palliative treatment was administered for low grade lymphoma and chronic lymphocytic leukaemia and when locally advanced stages or metastases were found in cancer of the breast, colon and rectum, stomach, pancreas, ovaries, urinary bladder, and in non-small cell lung cancer.

Most chemotherapy was administered on an outpatient basis, i.e., patients visited a hospital for treatment and were discharged the same day. More intensive treatment was given on an inpatient basis.

Resources and Costs

The direct cost for the care of cancer patients in 1996 was approximately 6.8 billion SEK, corresponding to approximately 6% of the total healthcare expenditure.

Specialized physicians and nurses are required to provide chemotherapy in an appropriate manner. This is important in order to deliver treatment on the appropriate indications and to treat side effects that may arise. Compared to, e.g., radiotherapy, chemotherapy requires no major investments in technology and drugs apart from qualified staff.

In Sweden, a large share of chemotherapy treatments, both curative and palliative, are delivered at specialized clinics for general and gynaecological oncology. Treatments are also given at departments of surgery, haematology, respiratory medicine, urology, and gynaecology.

The costs for chemotherapy are difficult to specify. They include not only the costs for cytostatic drugs but also costs for staff, facilities, and some equipment. Furthermore, there are costs for other drugs, e.g., to enhance the effects of cytostatic agents, to prevent certain side effects, and to influence various complications. Pre- and post-treatment investigations to assess the effects of treatment also involve some costs.

In 1998, the costs for cytostatic agents alone were 280 million SEK. Approximately one third of the costs were attributed to the group of newly developed cytostatic agents mentioned earlier, namely taxans. These and other new drugs substantially increased the costs for cytostatic drugs during the 1990s.

The drug costs to ameliorate the side effects of nausea and vomiting and the side effects that cytostatics have on formation of white blood cells in the bone marrow exceeded 110 million SEK. Combined, the costs for cytostatic agents and drugs to manage side effects are approximately 400 million SEK per year (1998), representing somewhat over 2% of the total drug costs.

The costs for cytostatic agents can vary substantially, from under 100 SEK to above 15 000 SEK per treatment cycle, and from around 200 SEK to over 100 000 SEK for a full treatment program extending 4 to 6 months. New cytostatic drugs cost substantially more than the older drugs. Most patients are treated with the older drugs since knowledge concerning their effects is better.

The costs for cytostatics should be assessed in relation to their effects or benefits. Studies of the cost effectiveness of chemotherapy, e.g., costs in SEK per quality-adjusted life-year gained, show that chemotherapy is at the same level and in many cases lower than well-established treatments for a range of different diseases. Knowledge about the cost effectiveness of treatment with cytostatics or other cancer treatment is, however, deficient.

Scientific Review

Several databases and other sources were used to identify relevant studies. The studies were classified and rank-ordered according to the weight of scientific evidence, using a system applied in other SBU projects. This process is described in Volume 1. Volume 1 also compares current practices of chemotherapy in Sweden wish the results from the scientific literature review. The literature review is presented in Volume 2. Each study was reviewed and graded by at least one member of the project group. The conclusions have been reviewed by a group of Swedish and international experts. The final conclusions concerning the weight of the evidence for each scientific study appear in the reference lists in the chapters of Volume 2.

The report is based on 1496 scientific studies covering 558 743 patients who received chemotherapy for the types of cancer studied. Most of the studies (1271) were comprised of meta-analyses, randomised prospective studies, or other prospective analyses with well-defined study protocols. Of these studies, 584 were found to contain high-grade scientific evidence. The quality was acceptable (moderate) in 543 studies and the remaining 369 studies were based on low-quality evidence.

The available scientific literature generally provides good information concerning the effectiveness of chemotherapy. In aggressive non-Hodgkin's lymphoma, Hodgkin's disease, and acute myelogenous leukaemia, chemotherapy may be the only treatment that cures the patient. In low grade non-Hodgkin's lymphoma and chronic lymphocytic leukaemia, chemotherapy has good palliative effects and can help patients to live well for many years without signs of disease. Chemotherapy in combination with surgery has the potential to cure breast cancer and colon cancer. In advanced or metastasized non-small cell lung cancer and cancer of the pancreas, colon and rectum, stomach, ovaries, urinary bladder, and breast, chemotherapy often alleviates symptoms. In many cases treatment also extends survival up to about 1 year.

The Scientific Evidence and Practices in Sweden

Approximately 2000 new cases of malignant lymphoma and 300 cases of acute myelogenous leukaemia are detected annually in Sweden. The literature clearly shows that these cancers are sensitive to both radiation and chemotherapy. In acute myelogenous leukaemia and advanced stages of aggressive non-Hodgkin's lymphoma, chemotherapy leads to cure in some patients. The cure rate varies among diseases and is high for the lymphomas called Hodgkin's disease and lower for acute myelogenous leukaemia. In early stages of lymphoma, radiotherapy also provides good treatment results, but a combination of chemotherapy and radiotherapy yields a lower risk for relapse. Combination treatment of this type is usually recommended since it probably leads to fewer late complications since a lower volume of tissue needs to be irradiated. In low grade non-Hodgkin's lymphoma and chronic lymphocytic leukaemia, chemotherapy has good effects on tumour cells but does not provide cure. However, the literature shows that these patients can live for an extended time with their disease, in some cases for decades.

High-dose chemotherapy has been tested in cases where the results from ordinary doses no longer provide sufficiently good effects. Treatment is given to only a limited number of patients, mainly in conjunction with scientific studies. The literature here is more deficient. Unequivocal evidence that high-dose treatment has positive effects is found only in a subgroup of patients with aggressive non-Hodgkin's lymphoma.

The survey of practice in Sweden shows that nearly all patients with aggressive non-Hodgkin's lymphoma and acute myelogenous leukaemia received chemotherapy aimed at cure. In low grade non-Hodgkin's lymphoma and chronic lymphocytic leukaemia, treatment was given to a smaller percentage of the patients and, as a rule, only when the disease showed signs of advancement. This agrees well with the results presented in the scientific literature.

Breast cancer is the most common type of cancer in females with approximately 5800 new cases per year. Chemotherapy and radiation therapy as complements to surgery reduce the risk for relapse. The literature on chemotherapy and breast cancer is extensive, and positive effects are well documented. In patients with a moderate or high risk for relapse of the disease following surgery, chemotherapy leads to improved survival and a chance for cure.

Chemotherapy in tumours that are already quite large on detection can often reduce them to a size where it is possible to operate or perform a breast conserving procedure. In metastasized breast cancer, palliative chemotherapy often allows the patient to live longer and feel better. Survival and life quality gains are also described after a second series of treatments when the effects of initial treatment are insufficient.

High-dose treatment has been tested both in patients at high risk for relapse following surgery and in relatively young patients with metastasized cancer where previous chemotherapy at normal doses showed good effects. However, obvious advantages with high-dose treatment have not been demonstrated.

The survey of practice showed that breast cancer is the diagnosis that accounts for most chemotherapy treatments. Treatment was delivered in line with well-documented effects presented in the literature, although some patients received a third and subsequent series of treatments. This is not supported in the scientific literature.

Colorectal cancer is one of the most common cancers, with approximately 5000 new cases annually.

The literature shows that chemotherapy for 6 months, as a complement to surgery, slightly increases the odds for cure in colon cancer that has metastasized to the lymph nodes. According to the survey of practice, such treatment was given to only a portion of the patients, which can be explained by that knowledge in this area was uncertain at the time.

Chemotherapy as a complement to surgery shows no confirmed effects on either rectal cancer or colon cancer that has not metastasized to the lymph nodes. In colon and rectal cancer that has metastasized to other organs, chemotherapy can extend survival by approximately 6 months and also improve the quality of life.

According to the survey, chemotherapy was given to a portion of the patients.

Positive effects from chemotherapy are also documented for a second series of palliative treatments when the effects of initial treatment are insufficient. According to the survey, a second series was given to only a few patients, which may be attributed to the lack of evidence at the time.

In Sweden, nearly 1200 people annually are diagnosed with gastric cancer and 900 with pancreatic cancer. Although these types of cancer are treated mainly by surgery, this is not possible in many cases. In advanced or metastasized cancer, chemotherapy may extend survival from a few months to half a year. The drugs may sufficiently influence cancer symptoms to improve the patients' quality of life. The effects are least apparent in pancreatic cancer.

The survey showed that only a limited percentage of patients received this type of palliative treatment.

Chemotherapy has not been shown to have positive effects as a complement to surgery for these tumours, and the survey showed that such treatment was not given.

Lung cancer is a common type of cancer in both women and men. Annually, 2800 new cases are diagnosed. The disease is found in various forms. The report includes only non-small cell lung cancer, which accounts for approximately 80% of all lung cancer in Sweden.

Chemotherapy as adjuvant treatment after surgery for early, non-small cell lung cancer has not been found to have any substantial effects. However, it appears that chemotherapy can offer minor survival gains when combined with radiotherapy or surgery for more advanced stages where metastasis is not present. In advanced or metastasized non-small cell lung cancer, chemotherapy can provide minor extensions in survival (1 to 3 months), and alleviation of symptoms has been well documented in some patients.

The survey shows that chemotherapy is given to only a small percentage of patients with non-small cell lung cancer, which agrees with the evidence in the literature concerning the limited effects of the treatment.

Approximately 900 new cases of ovarian cancer are diagnosed annually in Sweden. Ovarian cancer is usually detected at a late stage. Hence, it is often impossible to treat by surgery alone. Surgery and chemotherapy represent the most important treatment alternatives. The literature shows that advanced ovarian cancer often responds favourably to chemotherapy. Chemotherapy can cure or extend survival in a small percentage of patients, but the disease often returns even after initially successful treatment. The new taxans have shown better effects than earlier types of cytostatic agents, but longer observation times are needed for a more definitive assessment of the benefits of this new, expensive treatment method. Even following relapse, a second series of chemotherapy may provide limited, and often short-term, positive effects.

Positive effects from adjuvant chemotherapy at early stages have not been documented to date, probably because studies have been too small to show an expected effect.

The survey shows that a large percentage of patients received chemotherapy, probably motivated by the effects demonstrated in the literature. Some patients received chemotherapy even after the established treatment method was shown to have no effect, and this occurred at a higher rate than could be supported scientifically.

Approximately 2000 new cases of cancer in the urinary bladder are detected annually in Sweden. Surgery is mainly used to treat this type of cancer.

Adjuvant treatment with cytostatic drugs administered directly into the urinary bladder can reduce the rate of relapse in disease at an early stage. The survey of practice shows that a smaller number of patients received this adjuvant treatment. However, it is uncertain how large a percentage of the patients with early bladder cancer received such treatment.

The literature does not yet offer any clear evidence to show that adjuvant chemotherapy following surgery affects long-term survival in cancers that grow into and through the bladder wall. The survey showed that this treatment approach was seldom used.

Chemotherapy often reduces tumour size in metastasized, urinary bladder cancer. Some research suggests that survival is also extended, but the findings are poorly documented in the literature. The survey showed that chemotherapy was used to a very limited extent in these cases.

In summary, for all of the cancer types studied, a comparison between the scientific evidence and the results of the survey showed that both curative and palliative chemotherapy are delivered mainly in an evidence-based way in Sweden at the population level. For methodological reasons, one cannot exclude the possibility that in individual cases both overuse and underuse of chemotherapy may occur in palliative treatment.

Side Effects of Chemotherapy

Differences in the life processes between cancer cells and normal cells are minor. Therefore, it has not been possible to develop cytostatic agents that attack only the tumour cells and do not simultaneously affect normal cells. Chemotherapy involves a difficult balance between injuring and eliminating sufficient numbers of tumour cells while injuring as few normal cells as possible to avoid life threatening or severe side effects. Treatment by chemotherapy therefore involves negative effects on, e.g., blood formation, the gastrointestinal tract, liver, and kidneys. Other side effects include hair loss and injury to the nervous system. Some treatments can, e.g., affect the immune defense to the extent that life threatening conditions appear, which can lead to death in some cases.

The degree and frequency of side effects vary widely among different cytostatic drugs and among individuals. In some patients, some cytostatic combinations with a proven effect may have little impact while in other patients the same cytostatic combination may cause substantial and severe side effects.

The perspective on side effects depends on the goal of treatment. In situations where the goal is to cure, patients may accept severe side effects, particularly if the side effects are temporary. However, severe side effects should not be accepted in palliative treatment where the main goal is to improve the quality of life.

Impact on Health-Related Quality of Life

How patients experience treatment of cancer is an important issue, but one that has not been sufficiently addressed in the scientific literature. Studies on chemotherapy have seldom addressed the effects on health-related quality of life. The studies that have addressed patients’ quality of life and the side effects of treatment have focused mainly on methodological aspects, i.e., measuring and evaluating the patients’ situation and experiences in conjunction with cancer and chemotherapy. However, recent studies have shown that chemotherapy can favourably influence the quality of a patient’s life that has been lowered by disease. These studies mainly concern colorectal cancer, lung cancer, breast cancer, gastric cancer, and pancreatic cancer.

The methodology for measuring quality of life has not been sufficiently developed to provide simple, interpretable results. Nevertheless, many of the current standardized questionnaires can reliably reflect changes and illustrate international comparisons. These questionnaires, which the patients themselves can complete after brief instructions, cover well-defined key areas related to symptom burden, functional disability, and physical, psychological, and social well being. A major trend during the late 1990s was the increasing use of a smaller number of standardized life quality measures in clinical studies and trials. Thereby, new knowledge has been gained on clinical interpretation of the outcomes from these instruments and the additional value that patient-based information can provide.

Quality of life measurements have yet to be optimally utilized in designing clinical trials. A disadvantage has been that few studies are designed to give particular attention to quality of life as an outcome measure. It is also essential that future studies of different treatments attempt to assess the effects of components which are important to the patients’ perceived quality of life. The best approach for doing this has not yet been fully tested.

What Treatments Are Motivated?

In some diseases, the positive effects of curative and palliative chemotherapy are so apparent that the use of treatment does not need to be questioned even though controlled data may be somewhat deficient. This concerns, e.g., chemotherapy for various malignant lymphomas and acute myelogenous leukaemia. Good effects, even if they are not as dramatic, are also found in other cancers, e.g., ovarian cancer and breast cancer.

In other cancers or in some stages of cancer, the effects of chemotherapy may be well documented in the literature but nevertheless quite limited. Meta-analyses of randomised studies have made it possible to detect, with a high degree of certainty, small differences between newer and older treatment methods or between chemotherapy compared to no chemotherapy. The new and usually more effective anticancer treatments often have more side effects and cost more than older therapies. In these cases, it is substantially more difficult to determine whether the treatments should be included in routine health practice, as discussed further below.

Studies clearly show that patients demand treatments which have an impact on tumour growth. For that reason, but also because the treatment of cancer patients receives a high priority in health care, one can argue that new treatments should be immediately adopted when they are shown to be more effective. However, costs would increase rapidly in the near future if all treatments that are shown to provide some form of clinical benefit, even marginal benefit, would be routinely adopted.

The survey of practice shows that Swedish physicians are restrictive in offering palliative treatment for cancer in the colon and rectum, stomach, pancreas, lungs, and urinary bladder. This may be due either to having general reservations about the current state of knowledge or making reasonable interpretations of the scientific evidence, since the small treatment gains have been observed in clinical studies on selected patients (in whom treatment is probably more effective than in the general population of cancer patients).

The restrictive use may also depend on patients not choosing to be treated following open information on the limited effects of treatment and the potential side effects and reduction in quality of life. Clinical experience, however, shows that patients and their families request treatment even when physicians discourage it.

In situations where the effects of chemotherapy are minor, although well documented, it is impossible to determine if and when treatment should be routinely provided in Sweden today. Discussions between patients and physicians concerning side effects lead to a decision on what treatments should be given. Marginal advantages, such as limited opportunities for alleviating symptoms and improving the quality of life (and at most a few months additional life) may mean a lot to the individual patient. The extent to which these advantages offset, e.g., the major costs associated with the new treatments cannot be answered on a scientific basis, but constitute a question of values. Based on the literature reviewed, the report discusses recommendations concerning these situations. The recommendations can be viewed as a foundation for further discussions.

Future Development and Research

Advancements in knowledge about the characteristics of tumour cells and why tumour cells are resistant to cytostatic agents will continue to contribute toward improved outcomes and fewer side effects. A better understanding of tumour characteristics may improve the potential to predict treatment effects in the individual so that chemotherapy can be used only in patients who benefit from it.

Based on the scientific literature, the report reviews the projected future needs for chemotherapy in treating different cancers. The estimated increase in the incidence and mortality from cancer until year 2010 is based on the studies presented in the SBU report on radiotherapy (SBU Report No. 129, 1996). Despite some uncertainty in the prognoses, the data suggest that the need for both curative and palliative chemotherapy will increase. In a substantially longer perspective, improvements in other methods of treating patients with cancer may reduce the need for chemotherapy, but this is impossible to foresee at the present time.

SBU Conclusions

• Chemotherapy plays a role in both curative and palliative treatment in patients with the types of cancer addressed in this report. For some types of cancers, chemotherapy alone is decisive for successful treatment having a high probability of cure. In many other cancers, chemotherapy in combination with other treatment strategies, mainly surgery but also radiotherapy, plays an essential role for enhancing the potential for cure. For most cancers at advanced stages and involving metastases to other organs there is currently no curative treatment. Chemotherapy in these situations can, to various degrees, alleviate symptoms and extend survival.
• Side effects from chemotherapy are common. They may have substantial impact, e.g., on the number of blood cells, the gastrointestinal tract, liver, and kidneys and may cause hair loss and injury to the nervous system. The extent of side effects varies to a large degree among different individuals and treatments. This creates a difficult balance between the benefits and risks of treatment and is of particular concern in palliative treatment.
• At the time of the survey, chemotherapy in Sweden was practiced generally in accordance with the evidence presented in the scientific literature. Overutilization and underutilization appeared to be marginal.
• Since current cancer treatment is far from being totally successful, there is a great need for further research. Only about 10% of the treatments were given within the framework of clinical trials where an attempt was made to improve treatment effects. The percentage of patients included in such studies should be substantially increased, and the studies should also measure the effects on well being and quality of life.
• Some well-documented scientific studies have shown relatively small palliative effects of treatment with newer drugs or with older drugs in new therapeutic situations. According to the survey on practice, chemotherapy was delivered in these cases to a small percentage of the patients. If such treatment would be offered to all patients, it would require either investing more resources in health care or redistributing resources within the healthcare sector. It is essential to openly discuss the consequences of such options. Drug costs for chemotherapy in Sweden are currently about 400 million SEK per year. The cost in relation to benefit from chemotherapy can be perceived as high, but it is not notably different than the cost of treating many other diseases. The problem is not limited to chemotherapy. Many new methods may offer important benefits to individual patients, but represent a substantial increase in costs. The issue of what is reasonable in the relationship between cost and benefits is ultimately a question of values.