Introduction

Early palliative care (EPC) in patients with advanced solid cancers has shown several benefits, including increased survival.1

Recent studies have demonstrated that provision of EPC is useful also for patients undergoing stem cell transplantation and with acute myeloid leukaemia and recommended it as the new standard of care in this latter setting.2–5 This success has raised the need to identify other patients with haematological malignancies (HM) who may benefit from EPC.6 Patients with multiple myeloma (MM) represent a prime example of a population that could potentially benefit from this approach, as MM is incurable and affected patients typically experience high and unmet symptom needs.7 8

On these grounds, we examined the presence of quality indicators for palliative and end-of-life (EOL) care in a cohort of consecutive patients with MM receiving outpatient EPC and compared them with those of a cohort of patients receiving usual haematological care (UHC).

Patients and methodsStudy design and outcome

This is a retrospective cohort study of patients with MM who received EPC or UHC. We compared primary outcomes of quality indicators of PC and EOL care in patients who received EPC versus UHC. We also secondarily compared the number of treatments and overall survival (OS) between the EPC and UHC groups.

Population and interventions

From January 2011 to December 2020, all consecutive MM patients who had their treatment initiated at the study institution were considered eligible for this study. The demographic and clinical data extracted from hospital chart were reported in online supplemental material.

EPC was defined as integration of palliative care within 8 weeks from cancer diagnosis, as previously reported.1 4 After starting in that time frame from the diagnosis of MM, the EPC visits were delivered as previously described4 and reported in details in online supplemental material.

Usual haematological care

The frequency of UHC visits depended on the patient’s specific MM treatment plan. Patients undergoing UHC received standard haematological care with the supportive care measures instituted by the haematological team. The haematologists who provided UHC had not any training in palliative care.

Quality indicators for palliative and EOL care

We conducted electronically structured and comprehensive reviews of electronic hospital chart to determine the presence of quality indicators for palliative and EOL care for EPC and for UHC patients as previously reported4 and described in details in online supplemental material.

Statistical analysis

The statistical analysis is described in details in online supplemental material.

We used the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) cohort checklist when writing our report.9

Results

Overall 290 consecutive patients with MM were enrolled. Four were excluded because they had started treatment at other institutions before transitioning their care to the study institution. Of the remaining 286 patients, 55 received EPC and 231 received UHC (online supplemental figure S1). Median time of follow-up was 41 (range 1–117) months for EPC and 38 (range 1–118) for UHC patients. Clinical characteristics of the patients are detailed in online supplemental table S1.

Quality indicators of palliative care

Table 1 shows that EPC patients received a significantly higher number of quality indicators of PC compared with UHC patients, (mean 2.62±1.25 vs 1.12±0.95 (adjusted MR 2.18 (95% CI 1.75 to 2.73; p<0.001)) (online supplemental figure S1A,B).

Table 1

Quality indicators for palliative care in patients with multiple myeloma receiving EPC or UHC

In the EPC group, the median times from the first documented goals of care (GOC) discussion and ACP promotion to death were 162 days (range 3–1368) and 76 days (range 3–924), compared with 29 days (range 2–1595) for GOC in UHC patients, (p<0.001 and no calculable). The difference between patients of the two cohorts accessing home care services was not statistically significant (table 1).

In EPC patients but not in UHC patients, reduction in pain intensity improved significantly over time across the time points considered [mean Numeric Rating Scale (NRS) values 1.86±2.78 at T0, 1.03±2.24 at week 1 (p=0.01); 0.41±1.57 at week 4 (p=0.001)] (online supplemental table S2A).

Consistent with this, 40 (72.7%) out of 55 EPC patients received treatment with strong opioids compared with 129 (55.8%) UHC patients (adjusted OR 1.88 (95% CI 0.93 to 3.82; p=0.07)).

Mean duration of treatment with strong opioids was significantly longer in EPC than in UHC patients (p<0.001)) (online supplemental table S2B).

Quality indicators of EOL care

Of the entire study cohort, 115 patients died (22 in the EPC group, 93 in the UHC group) and formed the group in which we assessed quality of EOL care.

The analysis of the indicators of aggressiveness at EOL showed that, compared with UHC patients, EPC patients were less likely to receive aggressiveness at the EOL. However, these differences were not statistically significant (online supplemental table S3).

Number of treatment lines and OS

We calculated the number of treatment lines received by patients of the two cohort and whether differences in such a number could have influenced the survival. Compared with UHC patients, the EPC group received a lower mean number of treatment lines (1.53±0.77 vs 1.90±1.21, MR of 0.76 (95% CI 0.60 to 0.97; p=0.03)) and less frequently three and four or more lines of treatment (adjusted OR 0.33 (95% CI 0.13 to 0.86; p=0.02) and 0.11 (95% CI 0.01 to 0.89; p=0.03, respectively) (online supplemental table S1 and figure S3A).

Median OS was 5.30 years for EPC group and 5.46 years for UHC group (adjusted HR 0.92; 95% CI 0.56 to 1.51; p=0.7429) (online supplemental figure S3B).

Discussion

In this retrospective cohort study comparing patients with MM receiving EPC versus usual care, those receiving EPC had better pain control, with longer use of strong opioids; higher rates of symptom management; more frequent GOC discussions, earlier in the disease trajectory and a trend towards higher quality of EOL care.

Pain is the most frequent symptom and the most common cause of first PC consultation in patients with MM.7 10 A recent study reported that a PC intervention within 1 year from diagnosis may significantly contribute to its reduction.11 Our study provides novel insights by showing that in EPC patients, pain intensity was reduced already after 1 week, and throughout the follow-up, which was not the case for patients having received UHC. This finding underscores the relevance of integrating PC as early as possible in the disease trajectory of patients with MM.

Further primary clinical concerns of MM patients are represented by decreased emotional, physical and social functioning, coping with side effects and information needs.7 12 In our study, EPC patients were offered more frequently psychological support and obtained more information about the likely trajectory of the disease than UHC patients. These results, which are in line with previous observations in other cancer populations, indicate that EPC may provide a better coverage of MM patients’ needs and support its implementation also in MM routine practice settings.2–4

Guidelines in solid tumours and recent reviews in HM recommend engaging patients in conversations weighing explicitly the benefits and risks of continuing with disease-directed therapies and evaluating symptom-directed care.6 13 Our findings indicate that EPC patients were more frequently involved in discussions of prognosis and GOC and in the promotion of ACP than UHC patients.

Our study did not show a significant decrease of aggressiveness of care at the EOL, although EPC patients showed reduced ratios in 7 out of 7 indicators. However, it should be noted that less than 5% of EPC MM patients received anti-MM treatment in the last 14 days of life, when receipt of disease-directed therapies by less than 10% of patients with cancer in that timespan has been described as a condition associated with less aggressiveness of care.14 Moreover, the percentage of our EPC patients receiving anti-MM treatment in the last month of life is less than the 34% reported in a previous study of MM patients undergoing late palliative care, and suggests, again, that an earlier PC intervention may be associated with further reduction of aggressiveness at EOL.14

The trend for a reduced aggressiveness at EOL is further supported by the observation that significantly fewer EPC than UHC patients received three or more lines of treatment. Of note, prioritising symptom-directed care instead of disease-directed therapies did not negatively affect survival, as the OS of our EPC patients was similar to those of patients receiving UHC and similar to that reported in the literature.8

Our study has several limitations. The first one is the retrospective nature of the data. Also, incomplete data reporting may have underestimated the quality-of-care measures. Finally, being a single-centre study, results may have limited generalisability to other centres where trained supportive and palliative care teams may not be available. These limitations notwithstanding, our study represents one of the most comprehensive reports on patients with MM treated in a real-world setting and having received EPC.

In conclusion, our results suggest that EPC is feasible in patients with MM and results in better quality of care, including better management of pain, more psychological support, more frequent GOC and ACP discussions and a trend to reduced aggressiveness at the EOL, without negatively impacting survival. Our findings may also lay the groundwork for future prospective comparative studies in patients with MM.