Must patients with advanced cancer choose between a Phase I trial and hospice?
Phase I oncology trials offer no meaningful chance for direct medical benefit and they may prevent patients with advanced cancer from receiving palliative care in a hospice program. However, it is not known whether dual enrollment in a Phase I trial and hospice is feasible.
Five hundred thirty-four Phase I trials were identified in a national online database, of which 179 (34%) accepted patients with a life expectancy of less than 6 months. Of these, 50 were selected randomly. Their principal investigators were surveyed by fax, with follow-up telephone calls and e-mails. Ninety-two hospices were selected randomly from a national database. Surveys were conducted by telephone with intake coordinators. Principal investigators were asked whether patients enrolled in hospice could also enroll in their trials if they were eligible in all other respects. Hospice intake coordinators were asked whether a patient with advanced cancer who met hospice eligibility criteria could also enroll in a Phase I trial.
Surveys were completed by 45 of 50 principal investigators (90%) and by 89 of 92 hospices (97%). Although both groups were in favor of dual enrollment, principal investigators (41 of 45; 91%) were more likely to support dual enrollment than hospices (60 of 89; 67%; chi-square test, P = 0.004). Most hospices that did not support dual enrollment cited reasons that were based on concerns about payment or misunderstandings about the nature of Phase I trials.
Most hospices and Phase I principal investigators believe that eligible patients should be allowed to enroll simultaneously in hospice and Phase I trials. These results suggest that the choice between hospice and a Phase I trial is a false dilemma and that greater collaboration in this area is needed. Cancer 2002;95:1601–4. © 2002 American Cancer Society.
In recent years, there has been a dramatic increase in interest regarding palliative care for patients with advanced cancer.1 In this setting of growing interest, Phase I oncology trials have been placed in a difficult position. Agents and interventions that are tested in Phase I trials may have substantial side effects.2 In addition, they offer no meaningful chance of direct medical benefit,3 although patients believe that they do.4, 5 For these reasons, Phase I trials may appear to prevent advanced-stage cancer patients from receiving palliative care by offering them unrealistic hopes of benefit.6–8 This concern has led to suggestions that hospice care should be emphasized in the Phase I informed consent process as an alternative to enrollment.9
However, it is not clear whether patients who enroll in a Phase I trial must forgo or postpone access to hospice and the benefits hospice offers. Phase I trials are designed to determine a therapy's safety, not to test its efficacy. Therefore, patients with advanced cancer who desire palliative care without aggressive treatment might still logically enroll in a Phase I trial.
There are at least two reasons why dual enrollment deserves serious consideration. First, patients would have access to the interdisciplinary services that hospice offers, including aggressive pain and symptom management and counseling.10–12 Second, by facilitating dual enrollment, investigators and hospices can help patients to fulfill goals of altruism. Although altruism is not the primary reason that patients enroll in Phase I trials, it is nevertheless important to many patients.4, 5, 13, 14
The goal of this survey was to determine whether patients must choose between hospice and a Phase I trial or whether hospices and Phase I investigators believe dual enrollment is feasible.
MATERIALS AND METHODS
Hospices were identified using a national database provided by the National Hospice and Palliative Care Organization (NHPCO). This database is the most comprehensive listing of hospices available. It includes more than 4000 NHPCO members and nonmembers, with hospices that are certified by Medicare and those that are not. Ninety-two operating hospices were selected randomly from this database. The interviewer contacted each hospice's intake coordinator by telephone. After reading a scripted paragraph that described Phase I trials and emphasized that their goal is a determination of an agent's safety, the interviewer asked whether a patient with advanced cancer who was enrolled in that hospice could also enroll in a Phase I trial. All participants were asked to explain their answers. Because intake coordinators have varying degrees of autonomy in hospice admissions, responses were categorized as “unsure” if coordinators indicated that the decision would be made by someone else, such as a medical director. Coordinators were also asked to provide background information about their hospice and the characteristics of its patients.
From the online clinical trials database maintained by the National Cancer Institute,15 534 Phase I oncology trials were identified. Of these, 179 (34%) reported eligibility criteria that admitted patients with a prognosis of less than 6 months. Fifty investigators were selected randomly from this group. Investigators were contacted initially by fax, with a follow-up fax sent 1 week later. Using the same language that was used for the hospice survey, the investigator was asked whether a patient enrolled in hospice could enroll in the investigator's Phase I trial if the patient was eligible in all other respects. Nonrespondents were asked the same question by e-mail, if an e-mail address was known, or by telephone if the e-mail address was not known. Responses were coded as being in favor of dual enrollment if the investigator's only reservations were due to doubts that a hospice patient could meet the medical eligibility criteria of the trial.
Statistics were generated to describe hospice and Phase I trial characteristics. The chi-square test was used to compare the responses of investigators and intake coordinators. Stata software (Stata for Windows 5.0) was used for all statistical analysis. The University of Pennsylvania's Institutional Review Board approved this survey.
Surveys were completed by 45 of 50 principal investigators (90%) and by 89 of 92 intake coordinators (97%). The hospices surveyed closely resembled the population of hospices nationwide in terms of length of stay, profit status, and percent of patients with cancer.16 Table 1 summarizes the characteristics of these hospices and the Phase I trials of the surveyed principal investigators.
Table 1. Characteristics of Hospice and Phase I Trials
|Mean daily census (range)||52 (2–617)|
|Median length of stay (days)||31|
|Mean proportion of patients with cancer (%)||71|
|Phase I|| |
| Sponsor of trial|| |
| NIH||36 (80)|
| Pharmaceutical industry||7 (15)|
| Joint NIH and industry||2 (4)|
| Other||2 (4)|
|Minimum prognosis required for eligibility (weeks)|| |
| 4||3 (7)|
| 6||1 (2)|
| 8||13 (29)|
| 12||28 (62)|
|Type of malignancy|| |
| Solid tumor||32 (71)|
| Hematologic||10 (22)|
| Both||3 (7)|
|Intervention (overlapping categories)|| |
| Chemotherapy||24 (53)|
| Biologic (immunotherapy, gene therapy, fusion protein)||13 (29)|
| Other (Calcitriol, Dimesna)||4 (9)|
| Bone marrow transplant||3 (7)|
| Radiation therapy, radionuclides||2 (4)|
Most investigators (41 of 45; 91%) and hospices (60 of 89; 67%) said that eligible patients could enroll in hospice and a Phase I trial (Table 2). Investigators were more likely to approve of dual enrollment than were hospices (chi-square test, 8.38; P = 0.004).
Table 2. Opinions about Dual Enrollment in Hospice and Phase I Trials
|Hospice (n = 89)|| |
| Yes||60 (68)|
| Unsure||9 (10)|
| No||20 (22)|
|Investigators (n = 45)|| |
| Yes||41 (91)|
| No||4 (9)|
None of the investigators who were against dual enrollment (n = 4) provided an explanation. Intake coordinators who were against dual enrollment (n = 16) or who were unsure (n = 12) cited several reasons (range, 0–3). Fifteen said that any patient receiving curative therapy would not be eligible for hospice, six expressed concerns about payment by either Medicare (n = 3) or other payers (n = 3), and six gave no explanation. Only five coordinators described a philosophical objection or a policy against participation in research.
This study found that both intake coordinators and Phase I investigators were generally willing to consider the possibility of dual enrollment in hospice and Phase I trials. This suggests that patients in Phase I trials may be able to gain access to a range of services, including intensive nursing care, home health aide services, social work support, and counseling,12, 16 which are difficult to obtain through other channels. Phase I investigators may wish to consider ways in which they can facilitate hospice enrollment for those patients with severe symptoms and those with psychosocial needs who could benefit from the services hospice offers. In addition, hospices and Phase I trial centers may wish to explore collaborative relationships to provide hospice care more efficiently to those patients who might benefit.
These findings also suggest opportunities for education of hospice personnel and dissemination of information about Phase I trials. For instance, many hospices expressed misunderstandings about the goals of Phase I trials and the costs that hospices might incur for patients who enroll. Therefore, assurances about these aspects of Phase I trials are likely to further increase the willingness of hospices to consider dual enrollment.
Other obstacles to dual enrollment may be somewhat more difficult to overcome. For instance, some patients in Phase I trials may want to pursue aggressive life-sustaining therapy, which is beyond the scope of the hospice philosophy. Nevertheless, a variety of aggressive treatments are acceptable and a “do not resuscitate” order is not required for hospice enrollment.12 Although preferences regarding aggressive treatment may exclude some patients, it is likely that many more would be eligible.
A more serious obstacle, which was noted by some of the intake coordinators surveyed, is the Medicare regulations that govern reimbursement for hospice services. It is certain that Medicare regulations and the policies of fiscal intermediaries will be highly influential in determining whether dual enrollment is feasible. However, there is no clear Medicare position on this issue at the current time. Specifically, it is not clear whether an unproven agent being tested for safety in a Phase I trial would be treated by Medicare as a life-sustaining treatment that would make the Phase I subject ineligible for the hospice benefit. Nevertheless, these data suggest that there is strong support among both oncologists and hospice providers for dual enrollment. In light of these data, it seems reasonable for Medicare to take a formal position that is consistent with these opinions.
As these issues are clarified, it will also be important to develop collaborative relationships between Phase I centers and community hospice programs. To our knowledge, there are currently no such programs established. Bridge programs offer one option,17 but are not a substitute for the extensive services offered by hospice. Interdisciplinary palliative care services at research centers, like the one developed by Dr. Ann Berger at the National Institutes of Health, offer another option. However, they are limited to inpatient care. With education and closer relationships between academic medical centers and hospices, patients who enroll in Phase I trials can gain access to the services that hospice provides. These results suggest that the key stakeholders, i.e., Phase I investigators and hospices, believe that this goal is feasible.