Background: Graft-versus-host disease (GvHD) is an overactive systemic inflammatory response that can arise following allogeneic hematopoietic stem cell transplantation (HSCT). This condition occurs when the transplanted donor immune cells recognize the recipient’s tissues as foreign and trigger an immune response against them. The ocular surface (eyelids, conjunctiva, meibomian glands, lacrimal glands, and cornea) is particularly involved in GvHD, and its response to existing treatments, including potent immunosuppressants and new targeted therapies, is undesirable, with such treatments often being ineffective. Human allogeneic umbilical cord blood platelet lysate stands out as a potent adjunct to conventional therapies for ocular surface disorders related to severe Dry Eye Disease. This study aimed to evaluate the safety and efficacy of umbilical cord blood platelet lysate eyedrops for the treatment of severe ocular surface disorders in graft-versus-host disease patients who have received previous unsuccessful treatments. Methods: This study was a prospective, non-comparative, interventional case series study involving 22 patients (10 females and 12 males) aged 25–46 years with severe ocular surface disorders that were unresponsive to standard treatments. The GvHD patients were categorized based on the severity of their ocular surface disorders into three groups: Group I: five patients with severe Dry Eye Disease and filamentary keratitis; Group II: eight patients suffering from severe blepharo-kerato-epitheliopathy; Group III: nine patients with corneal ulcers. Fresh umbilical cord blood (UCB) was obtained from healthy donors and subjected to centrifugation using a novel PRP preparation kit provided by Sciacca (AG) Cord blood bank, Italy in a one-step process. In all groups, the outcomes before and after treatment were evaluated by means of the OSDI (Ocular Surface Disease Index), SANDE (Symptom Assessment in Dry Eye) questionnaire, VAS (Visual Analogue Scale), slit lamp examination, Esthesiometry, Lissamine Green Staining, the NIBUT (Non-Invasive Break-Up Time) and BUT, fluorescein staining with digital photography and Oxford classification, the Schirmer Test, the Best Corrected Visual Acuity (BCVA), and Meibography. In Group III at each evaluation time, the size of the ulcer and its relative reduction compared to the baseline size were recorded. Clinical variables, such as corneal inflammation, conjunctivalization, corneal neovascularization, or pain, were also considered individually. Results: We observed a significant improvement in the SANDE, VAS, and OSDI scores; Schirmer Test; BUT; BCVA; and Oxford classification after treatment with allogeneic cord blood serum eyedrops. Nevertheless, pain and inflammation reduced markedly over time until complete healing in all cases. The mean reduction in the ulcer surface area (compared to baseline values) was significantly higher at all assessment points (p = 0.001 for day 7 and p < 0.001 for subsequent time points every 30 days for 90 days). At the last check-up (after 90 days of treatment), the number of ulcers (Group III, nine patients) with a reduction in size of greater than 50% was eight (88.8%), of which seven ulcers were completely healed. None of the patients experienced treatment-related local or systemic adverse events. In this study, using a relatively large number of cases, we demonstrated that the use of umbilical cord blood platelet lysate eyedrops is a safe, feasible, and effective curative approach for severe ocular surface disease in patients with GvHD. Conclusions: Our pilot study highlights the remarkable effectiveness of allogeneic cord blood serum eyedrops in patients with severe ocular surface disorders following GvHD who have shown an inadequate response to the usual treatments. It is mandatory to design future studies on the efficacy of this therapeutic approach for acute ocular, mucosal, and cutaneous GvHD.