The French Society for Pharmacy Oncology's commentary on the article by Iannopollo M.,published in April 2026 and titled “Rethinking centralised oncology pharmacy practice in the era of flat-dose immunotherapy,subcutaneous formulations and visit-administration day separation: A conceptual analysis and reform proposal”

Iannopollo's article, published in 2026, discusses the production of injectable anticancer drugs by hospital pharmacists. While the manuscript raises important questions regarding the evolution of oncology pharmacy practice, the French Society for Oncology Pharmacy (SFPO) would like to offer a critical perspective on several of its core assumptions and conclusions.

First, the author does not provide a comprehensive account of the prescription review conducted by hospital pharmacists prior to the preparation of injectable anticancer drugs. In reality, this in-depth analysis includes verification of the indication (type of cancer, validated protocol, reimbursement), compliance with guidelines (institutional protocols, international guidelines), and verification of the treatment regimen (days of administration, cycles, sequences), in addition, of course, to the dosage, which, with fixed-dose formulations, is less prone to error, except for patients with extreme body weight. Recent laboratory results (complete blood count, renal and hepatic function tests, etc.) and the patient's tolerance of previous treatment cycles (adverse effects) are also taken into account prior to pharmaceutical validation of injectable anticancer drugs, included for flat-doses.^1,2 These immunotherapies are often used in combination with other cancer treatments, therefore, the overall appropriateness of the prescription is verified. The author affirms that « procedural non-differentiation between pharmacologically non-equivalent preparations is the primary structural inefficiency of the current model », but there is no data to support this hypothesis. Thus, the absence of treatment dose calculations does not eliminate clinical complexity or the need for centralized pharmaceutical preparation.

Secondly, the premise of this article is that the apparent pharmacological simplification (flat-dose, subcutaneous injection) would automatically lead to organizational simplification. The subcutaneous route is not a new concept (cytarabine, bortezomib, rituximab and trastuzumab were the first). It should be emphasized that substantial organizational transformations have already taken place within hospitals, particularly regarding the containment of nursing workforce expenditures. ³ In 2022, the World Health Organization issued a warning: the shortage of healthcare workers was a “ticking time bomb.” New estimates for the entire European region suggest there will be a shortage of nearly one million healthcare workers by 2030. ⁴ In this context, transferring medication preparation activities from pharmacy departments to clinical wards would not only increase the workload of nursing staff but also create a structural mismatch with existing staffing models, as current nursing budgets are not designed to absorb such additional technical responsibilities. Addressing this gap would require significant and sustained recruitment of nurses, which appears neither realistic nor aligned with current workforce constraints in many healthcare systems. However, many pharmacies specializing in oncology already prepare anticancer drugs in advance, including flat dose and syringes for subcutaneous injection, which allows healthcare professionals to devote more time to patient care. ⁵ In a survey conducted in France in 2021, the SFPO found that 65% of respondents (n = 73) used advance preparation and 27% (n = 28) used standardized doses for anticancer drugs. ⁶ So, the time required to prepare subcutaneous forms by hospital pharmacists is not a very convincing argument, given that this preparation can be done in advance. The impact of the time required for nurses on the wards to prepare subcutaneous injections has not been assessed in terms of organizational efficiency or nurse and patient satisfaction. Moreover there is a scarcity of information about the occupational health and safety risks associated with handling monoclonal antibody.^7,8 Most of these drugs do not contain antimicrobial preservatives, so an aseptic technique must be followed during preparation.

Consequently, the SFPO recommends maintaining the centralized preparation of anticancer drugs, including subcutaneous monoclonal antibodies, by redefining workflows and investing in preparation assistance systems or automation. The growing implementation of robotic compounding systems enables a substantial reduction in pressure on pharmacy technician resources while maintaining a very high level of safety. Compared with ward-based preparation, centralized automated compounding offers superior control over aseptic conditions, reduces the risk of microbial contamination, and improves overall process standardization. In addition, these technologies contribute to economic optimization by enhancing productivity and limiting drug waste.^9,10 Maintaining consistent procedures across anticancer drug classes may reflect a deliberate safety strategy rather than inefficiency. The proposal to decentralize certain activities, particularly for subcutaneous formulations, should therefore be approached with caution, as it may reintroduce variability at the point of care. Finally, the central issue may not lie in centralized pharmacy compounding itself, but rather in the availability of ready-to-use pharmaceutical forms provided by manufacturers, such as prefilled syringes for subcutaneous injection. These formulations would allow for enhanced traceability, improved pharmaceutical safety, and streamlined nominal dispensing on a day-to-day basis. Moreover, they could significantly alleviate pressure on human resources, enabling pharmacy teams to focus on high value-added activities, including the management of innovative anticancer therapies, clinical trials, and advanced therapy medicinal products, whose number and complexity continue to increase in oncology practice.

Thirdly, the economic analysis is incomplete. Many treatment vials (including fixed-dose antibody vials) are overfilled, allowing the residual product to be used to generate savings. These savings are significant due to the high unit price. This line of reasoning contrasts the cost of preparation with the cost of the medication, but overlooks the cost of risk—which is, after all, central to healthcare. Although fixed-dose regimens may facilitate reassignment of unused preparations under certain conditions, this remains strictly feasible within a controlled environment that ensures aseptic preparation and full traceability. Such reassignment cannot be safely achieved in decentralized clinical settings. While it is true that ward-based preparation may reduce the risk of drug wastage in case of last-minute treatment cancellations, this benefit must be carefully balanced against the increased risks associated with non-controlled environments. In several countries, including France, subcutaneous immunotherapies are increasingly being directed toward home-based care pathways. In this context, preparation at the patient's bedside or in non-aseptic environments further increases the risk of microbial contamination. This not only raises safety concerns for patients but also heightens the likelihood of product loss, particularly as vial overfill cannot be effectively utilized outside controlled pharmacy settings.

Finally, the argumentative structure of this conceptual analysis is unevenly supported by the bibliography, with several central claims resting on references whose nature or scope does not correspond to the evidentiary weight they are asked to bear. The assertion that pharmaceutical validation errors increase under time pressure is attributed to Kaushal, Shojania and Bates, a systematic review whose actual object is the efficacy of computerized physician order entry and clinical decision support systems in reducing medication errors. Time pressure is neither a study variable nor an outcome measure in that work; the citation exploits thematic proximity to medication safety without engaging the specific mechanism invoked, and would not survive scrutiny in peer review. The claim that ward-based nurses prepare subcutaneous formulations three times faster than pharmacy staff rests on a congress oral presentation [Cappuzzo, ELCC 2024, Prague], a non-peer-reviewed, non-published and non-reproducible source that cannot bear the organizational weight of a structural reform recommendation. The normative grounding of the D/D + 1 model partly relies on an internal institutional procedure [ASL AL, PG.ONC.001, 2022], a non-indexed, non-peer-reviewed document whose evidentiary value is descriptive and non-generalizable. Most critically, the claim of international convergence toward simplified workflows for low-complexity preparations invokes SFPO recommendations on “préparations à faible complexité” [SFPO, 2022], a reference that corresponds to no published, indexed or publicly available document issued by our society. This is not an isolated concern: ¹¹ cited as Kastrup M, Isermeyer L, Dähne A, et al., “Robotic preparation of chemotherapy: safety and efficiency outcomes from a German multicentre evaluation,” J Oncol Pharm Pract 2023; 29: 502–510 — could likewise not be identified in any bibliographic database, including PubMed, nor in the corresponding volume of the journal. The coexistence of at least two unverifiable references, one attributed to a named scientific society and one to a peer-reviewed journal, raises serious questions about the reliability of the bibliographic verification process as a whole. ¹¹

To conclude, the production of anticancer treatment is embedded in a quality improvement dynamic that cannot and must not be reversed. The Deming cycle, applied to the chemotherapy circuit, has enabled major documented advances including centralization, pharmaceutical validation, traceability, and secured interfaces, now further strengthened by robotic compounding, automated controls, and artificial intelligence. ¹² These innovations enhance safety and efficiency without displacing pharmaceutical expertise; they refocus it where it matters most. The growing posological complexity of modern anticancer agents, illustrated by nivolumab approved at 240 mg Q2 W, 480 mg Q4 W, and 360 mg Q3 W, or pembrolizumab at 200 mg Q3 W and 400 mg Q6 W, demonstrates that pharmacological proximity does not imply operational equivalence, and that the risk of schedule confusion is real, even where under-reported. Proposing to differentiate circuits on the basis of criteria that are neither evaluated nor supported by the literature would amount not to advancing this wheel, but to turning it back. Maintaining consistent procedures across anticancer drug classes reflects a deliberate safety strategy, not inefficiency. As the complexity and volume of anticancer therapies continue to grow, the appropriate response is to strengthen safety systems and invest in smarter workflows, not to weaken them, for the benefit of patients and healthcare professionals alike.