Ninety-three patients were treated with M-PHP between January 2019 and April 2023, from which 45 patients consented to participate prior to the first M-PHP. Thirty-three patients filled in all four questionnaires and were eligible for analyses. From this group, nine patients received combination treatment with M-PHP and immunotherapy as part of a randomized phase II trial. These patients were excluded from the analyses as all study data were under embargo until completion of the trial, resulting in a study cohort of 24 patients (Fig. 1).

Flowchart of patients. M-PHP Percutaneous hepatic perfusion with melphalan

The study population consisted of 14 males and 10 females. The median age was 63.0 years (range 47–74). Twelve patients underwent enucleation as treatment for the primary tumor, 10 patients were treated with radiotherapy. Nine patients received prior treatment for hepatic metastases, consisting of ablation (n = 5), surgical resection (n = 2), or ablation of a liver metastases with cryoablation of a lesion in the sacrum (n = 1). One patient received systemic treatment with immunotherapy prior to treatment with M-PHP. All patients presented with metachronous metastases and multifocal disease (Table 1).

A total of 96 questionnaires were analyzed of the 24 patients. The scores per scale (global health status, symptom scales, and functional scales) are presented in Table 2 and Fig. 2.

Scale scores EORTC QLQ-C30 as median with interquartile ranges. For global health status scale and functional scales, a higher score indicates better quality of life or functional status. For the symptom scales, a higher score indicates more symptoms. P-values were based on the comparison between score at baseline and on day 21

The median GHS score was 83 at baseline. Compared to baseline, on day 2/3 post M-PHP, the median score decreased to 58 (p < 0.001). Similarly, the median score on day 7 after M-PHP was significantly lower compared to baseline (67; p < 0.001). On day 21, the score restored to baseline value 83 (p = 0.034). One patient had a markedly lower score compared to the rest of the group: a decrease in absolute GHS score of 50 points from 83 at baseline to 33 on day 21. The deterioration in GHS score could be attributed to a grade 3 post-procedural hemorrhage in the groin, for which transfusion of two units of red blood cells followed.

All symptom scores started at median 0, except for the fatigue score (Table 2 and Fig. 2). On day 2/3 after M-PHP, patients gave a median score of 17 for nausea/vomiting (p < 0.001) and a median score of 33 for pain (p = 0.005), insomnia (p = 0.001), and appetite loss (p < 0.001). The scores for dyspnea, constipation, diarrhea, and financial difficulties did not show statistically significant change compared to baseline (p > 0.0167 for all). The median fatigue score was 6 at baseline and increased to 50 at day 2/3 (p < 0.001). On day 7 after treatment, fatigue (44; p < 0.001), nausea/vomiting (0; p = 0.014), dyspnea (0; p = 0.009), insomnia (33; p = 0.010), and appetite loss (0; p = 0.004) were higher compared to baseline. The median scores for pain, constipation, diarrhea, and financial difficulties restored to baseline on day 7 (p > 0.0167 for all). On day 21, only fatigue had a higher median score compared to baseline (33; p = 0.002), while all other symptom scores had returned to baseline (p > 0.0167 for all).

All functional scores started with median score 100, except for the emotional functioning score (Table 2 and Fig. 2). Compared to baseline, physical functioning (57; p = 0.001), role functioning (33; p < 0.001), cognitive functioning (83; p = 0.001), and social functioning (67; p < 0.001) showed a significant decline on day 2/3 after M-PHP. The median emotional functioning score was 79 at baseline and 75 on day 2/3 (p = 0.387). On day 7 after M-PHP, physical functioning (73; p < 0.001), role functioning (50, p < 0.001), and social functioning (67; p = 0.007) scored significantly lower compared to baseline. Emotional functioning (79; p = 0.774) and cognitive functioning (100; p = 0.046) restored to baseline values. On day 21, the scores for physical (87; p = 0.003) and role functioning (67; p = 0.001) remained significantly lower compared to baseline, while the other scores restored to baseline values (p > 0.0167 for all).

The M-PHP procedures had a median duration of 3.2 h (range 2.3–4.8 h). The median melphalan dose for the whole cohort was 220 mg (range 153–220 mg; Table 3).

Two patients experienced periprocedural complications. One patient experienced hypothermia, hypotension, metabolic acidosis, and cardiac complications (atrial fibrillation, bradycardia, and ST-depressions) and needed to stay intubated after the procedure. The patient recovered within hours after the procedure and could be discharged on day 3 without any sequelae. The other patient had a pseudoaneurysm of the common femoral artery treated with thrombin injection (Table 3).

All AEs within 30 days after M-PHP were registered. All patients experienced post-procedural AEs, resulting in a total of 134 reported AEs. Twenty-one (15.7%) of all post-procedural AEs were grade 3 or 4 (Table 4). Nine patients experienced the aforementioned total of 21 grade 3 or 4 AEs, of which 16 were hematological AEs. The median GHS score of these patients was 67 on day 21. This score was not significantly different compared to patients who experienced low-grade AEs (median GHS score 83, p = 0.174). Asymptomatic grade 1 or 2 anemia (n = 16), thrombopenia (n = 13), and increased lactate dehydrogenase (LDH), alanine aminotransferase (ALT), and gamma-glutamyl transferase (GGT) (n = 11) were the most frequent AEs. Patients also reported transient nausea (n = 6), fatigue (n = 5), and alopecia (n = 5). Grade 3 or 4 leukopenia and thrombocytopenia were most frequently reported (n = 6 each). No deaths occurred during treatment period.