Improvement in corneal sensitivity at the overall Follow-up Month 24 and 30 was assessed as a binary goal attainment variable (Yes/No), based on the subset of patients who did not achieve an improvement in corneal sensitivity at Week 8 of the NGF0120. Improvement was defined as a change from Baseline > 0 cm in the study eye. Corneal sensitivity was measured using the Cochet-Bonnet aesthesiometer in the central NEI zone before the instillation of any anesthetic or dilating drops. The filament was extended to 6 cm and retracted in 0.5 cm increments until the patient reported sensation upon contact. Higher values indicate better sensitivity. Please note that patients without a Yes/No response available (=missing data) are not considered in the analysis. More in details, these missing patients are 1 at month 30 only.

Corneal sensitivity at the overall Follow-up Month 24 and Month 30 measured (in cm) in the qualifying NEI (National Eye Institute) zone of the study eye using the Cochet-Bonnet aesthesiometer before the instillation of any anesthetic or dilating drops.

Improvement was defined as a change from Baseline > 0 cm in the study eye. Higher values indicate better sensitivity.

Change in Schirmer I test scores was assessed as a continuous outcome in the study eye from Baseline and at Week 8 of the NGF0120 to Month 24 and Month 30.

The Schirmer I test was conducted on unanesthetized eyes using standardized paper test strips placed in the lower temporal lid margin of each eye. After 5 minutes, the strip was removed, and the length of the moistened area (in millimeters) was recorded.This test measures aqueous tear production. Any change in mean score >0 was considered an improvement in dry eye. Higher scores indicate better tear production, while lower values are associated with more severe dry eye. A positive change from Baseline was interpreted as improvement.

Change in Tear Film Break-Up Time was evaluated in the study eye. TFBUT was measured in seconds using fluorescein dye under cobalt blue illumination. After fluorescein instillation, the patient blinked once or twice, then kept the eye open; the time to first dry spot was recorded. Two values were averaged, or three if differing by >2 seconds. TFBUT values ranged from 0 seconds (tear film instability) to ≥10 seconds (normal stability). Values <5 seconds suggest clinically relevant tear film dysfunction. A positive change from Baseline or Week 8 (>0 seconds) was interpreted as improvement in tear film stability.

Best Corrected Distance Visual Acuity (BCDVA) at the overall Follow-up Month 24 and Month 30 was assessed in the study eye using the ETDRS visual acuity chart at a distance of 4 meters (13.1 feet), prior to any administration of anesthetic or mydriatic eye drops. Results were recorded as the number of letters correctly identified on the chart. This outcome evaluated the number and percentage of patients who gained at least 15 letters in BCDVA from DEFENDO Baseline and week 8, at Month 24 and Month 30. A gain of ≥15 letters is considered a clinically meaningful improvement corresponding to approximately three lines on the ETDRS chart. The endpoint was analyzed as a binary variable (Yes/No).

Best Corrected Distance Visual Acuity (BCDVA) at Month 24 and Month 30 was assessed in the study eye using the ETDRS visual acuity chart at 4 meters (13.1 feet), before any anesthetic or mydriatic eye drops or ocular procedures. Scores were recorded in logMAR units, ranging from -0.3 (better visual acuity) to 1.0 (poorer acuity). Higher logMAR values indicate worse vision. A negative change in logMAR (i.e., <0) indicated an improvement in visual acuity. Mean values and standard deviations were summarized descriptively at each timepoint.

"Impact of Dry Eye on Everyday Life" (IDEEL) is a 57-item questionnaire assessing dry eye impact (and consequently QoL), composed by 6 domains:

• Daily Activities; Higher score = less impact
• Emotional Impact: Higher score = less impact on emotions
• Impact on Work: Higher score = less impact on work
• Symptom Bother: Higher score = greater symptom bother
• Treatment Bother: Higher score = less treatment-related bother
• Treatment Effectiveness: Higher score = greater satisfaction with treatment effectiveness For 5 of 6 domains (Symptom Bother), higher scores on a 0-100 scale indicated improved quality of life relative to dry eye symptoms. For the domain of "symptom bother", a higher score indicated greater symptom bother. For this reason no total score, but only subscores, is calculated.

The EQ-5D-5L is a standardized questionnaire what measures the health status in 5 areas-mobility, self-care, usual activities, pain/discomfort, and anxiety/depression-with a rating scale of 1 (no problems) to 5 (extreme problems), where higher scores indicated worse problems.

The questionnaire also asks the patient to rate their current health ("how good or bad your health is today") on a scale from 0 (worst health you can imagine) to 100 (best health you can imagine). On a 0-100 scale, a higher "your health today" score indicated better health. The EQ-5D-5L questionnaire was self-administered by the patient before any ophthalmic procedures at either visit.

Please note that the severity level reported corresponds to the maximum experienced by the patient (for example, if a patient has one mild AE and one moderate AE, they are counted in the moderate category).

An AE was defined as any untoward medical occurrence in a patient or clinical investigation patient administered a pharmaceutical product that did not necessarily have a causal relationship with this treatment. An AE could therefore be any unfavorable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of the medicinal (investigational) product, whether or not related to the study product.

AEs by severity (categorized as mild, moderate and severe) and relatedness (categorized as related and not related ) are considered - per each single subject - just once, at the greater severity and closest relationship to treatment. Considering that no study IMP was administered and that all standard of care were permitted, AEs must be read as related to any standard of care administered during the FU study.

Slit-lamp biomicroscopy was performed in the study eye at Baseline and Week 8 of the NGF0120 study, and at Month 24 and 30. The anterior segment was evaluated under magnification using standard slit-lamp illumination. Parameters included eyelid edema and erythema, lashes, conjunctiva edema and bulbar erythema, cornea edema, endothelial and epithelial changes, lens, sclera, iris, and anterior chamber cells and flare. Findings were assessed visually as normal, abnormal not clinically significant, or abnormal clinically significant, based on clinical judgment and standard ophthalmologic criteria. Only the participants reporting normal and abnormal clinically significant results are shown (along with the percentage vs the "Number Analyzed", for each timepoint) The number of participants with abnormal not clinically significant results can be derived as difference between the sum of the said two categories and the "Number Analyzed", per each parameter and timepoint

Corneal epithelial healing at the overall Follow-up Month 24 and Month 30 was summarized as a binary goal attainment variable (Yes/No) based on the subset of patients who did not achieve corneal epithelial healing at Week 8 of the NGF0120, as determined by the Central Reading Center (CRC). Healing was defined as the absence of persistent epithelial staining abnormalities related to disease.

Epithelial healing took into consideration the Baseline epithelial staining patterns of each patient and their evolution over time.The persistence (or worsening) of a staining pattern in the same corneal area was considered a failure in terms of healing.

The Full Analysis Set (FAS) included all enrolled patients who attended either the Month 24 or Month 30 study visit or both in this extension study. The FAS was used for both efficacy and safety analyses.

Please note that while at the Month 24 visit both patients had available data, at the month 30 only one of the two patients had available data.

Improvement in corneal sensitivity at the overall Follow-up Month 24 and 30 was assessed as a binary goal attainment variable (Yes/No), based on the subset of patients who achieved an improvement in corneal sensitivity at Week 8 of the NGF0120. Improvement was defined as a change from Baseline > 0 cm in the study eye. Corneal sensitivity was measured in the central National Eye Institute (NEI) zone using the Cochet-Bonnet aesthesiometer, before the instillation of any dilating or anesthetic drops. The filament was extended to its full length (6 cm), then retracted in 0.5 cm increments until the patient reported feeling contact. The filament length at which sensation was reported was recorded. Higher values indicate better sensitivity. Please note that patients without a Yes/No response available (=missing data) are not considered in the analysis. More in details, these missing patients are 4 at month 24 and 3 at month 30.

The Full Analysis Set (FAS) included all enrolled patients who attended either the Month 24 or Month 30 study visit or both in this extension study. The FAS was used for both efficacy and safety analyses.

Improvement in corneal sensitivity at the overall Follow-up Month 24 and 30 was assessed as a binary goal attainment variable (Yes/No), based on the subset of patients who did not achieve an improvement in corneal sensitivity at Week 8 of the NGF0120. Improvement was defined as a change from Baseline > 0 cm in the study eye. Corneal sensitivity was measured using the Cochet-Bonnet aesthesiometer in the central NEI zone before the instillation of any anesthetic or dilating drops. The filament was extended to 6 cm and retracted in 0.5 cm increments until the patient reported sensation upon contact. Higher values indicate better sensitivity. Please note that patients without a Yes/No response available (=missing data) are not considered in the analysis. More in details, these missing patients are 1 at month 30 only.

The Full Analysis Set (FAS) included all enrolled patients who attended either the Month 24 or Month 30 study visit or both in this extension study. The FAS was used for both efficacy and safety analyses.

Corneal sensitivity at the overall Follow-up Month 24 and Month 30 measured (in cm) in the qualifying NEI (National Eye Institute) zone of the study eye using the Cochet-Bonnet aesthesiometer before the instillation of any anesthetic or dilating drops.

Improvement was defined as a change from Baseline > 0 cm in the study eye. Higher values indicate better sensitivity.

The Full Analysis Set (FAS) included all enrolled patients who attended either the Month 24 or Month 30 study visit or both in this extension study. The FAS was used for both efficacy and safety analyses.

Change in Schirmer I test scores was assessed as a continuous outcome in the study eye from Baseline and at Week 8 of the NGF0120 to Month 24 and Month 30.

The Schirmer I test was conducted on unanesthetized eyes using standardized paper test strips placed in the lower temporal lid margin of each eye. After 5 minutes, the strip was removed, and the length of the moistened area (in millimeters) was recorded.This test measures aqueous tear production. Any change in mean score >0 was considered an improvement in dry eye. Higher scores indicate better tear production, while lower values are associated with more severe dry eye. A positive change from Baseline was interpreted as improvement.

The Full Analysis Set (FAS) included all enrolled patients who attended either the Month 24 or Month 30 study visit or both in this extension study. The FAS was used for both efficacy and safety analyses.

Change in Tear Film Break-Up Time was evaluated in the study eye. TFBUT was measured in seconds using fluorescein dye under cobalt blue illumination. After fluorescein instillation, the patient blinked once or twice, then kept the eye open; the time to first dry spot was recorded. Two values were averaged, or three if differing by >2 seconds. TFBUT values ranged from 0 seconds (tear film instability) to ≥10 seconds (normal stability). Values <5 seconds suggest clinically relevant tear film dysfunction. A positive change from Baseline or Week 8 (>0 seconds) was interpreted as improvement in tear film stability.

The Full Analysis Set (FAS) included all enrolled patients who attended either the Month 24 or Month 30 study visit or both in this extension study. The FAS was used for both efficacy and safety analyses.

Best Corrected Distance Visual Acuity (BCDVA) at the overall Follow-up Month 24 and Month 30 was assessed in the study eye using the ETDRS visual acuity chart at a distance of 4 meters (13.1 feet), prior to any administration of anesthetic or mydriatic eye drops. Results were recorded as the number of letters correctly identified on the chart. This outcome evaluated the number and percentage of patients who gained at least 15 letters in BCDVA from DEFENDO Baseline and week 8, at Month 24 and Month 30. A gain of ≥15 letters is considered a clinically meaningful improvement corresponding to approximately three lines on the ETDRS chart. The endpoint was analyzed as a binary variable (Yes/No).

The Full Analysis Set (FAS) included all enrolled patients who attended either the Month 24 or Month 30 study visit or both in this extension study. The FAS was used for both efficacy and safety analyses.

Best Corrected Distance Visual Acuity (BCDVA) at Month 24 and Month 30 was assessed in the study eye using the ETDRS visual acuity chart at 4 meters (13.1 feet), before any anesthetic or mydriatic eye drops or ocular procedures. Scores were recorded in logMAR units, ranging from -0.3 (better visual acuity) to 1.0 (poorer acuity). Higher logMAR values indicate worse vision. A negative change in logMAR (i.e., <0) indicated an improvement in visual acuity. Mean values and standard deviations were summarized descriptively at each timepoint.

The Full Analysis Set (FAS) included all enrolled patients who attended either the Month 24 or Month 30 study visit or both in this extension study. The FAS was used for both efficacy and safety analyses.

"Impact of Dry Eye on Everyday Life" (IDEEL) is a 57-item questionnaire assessing dry eye impact (and consequently QoL), composed by 6 domains:

• Daily Activities; Higher score = less impact
• Emotional Impact: Higher score = less impact on emotions
• Impact on Work: Higher score = less impact on work
• Symptom Bother: Higher score = greater symptom bother
• Treatment Bother: Higher score = less treatment-related bother
• Treatment Effectiveness: Higher score = greater satisfaction with treatment effectiveness For 5 of 6 domains (Symptom Bother), higher scores on a 0-100 scale indicated improved quality of life relative to dry eye symptoms. For the domain of "symptom bother", a higher score indicated greater symptom bother. For this reason no total score, but only subscores, is calculated.

The Full Analysis Set (FAS) included all enrolled patients who attended either the Month 24 or Month 30 study visit or both in this extension study. The FAS was used for both efficacy and safety analyses.

The EQ-5D-5L is a standardized questionnaire what measures the health status in 5 areas-mobility, self-care, usual activities, pain/discomfort, and anxiety/depression-with a rating scale of 1 (no problems) to 5 (extreme problems), where higher scores indicated worse problems.

The questionnaire also asks the patient to rate their current health ("how good or bad your health is today") on a scale from 0 (worst health you can imagine) to 100 (best health you can imagine). On a 0-100 scale, a higher "your health today" score indicated better health. The EQ-5D-5L questionnaire was self-administered by the patient before any ophthalmic procedures at either visit.

Please note that the severity level reported corresponds to the maximum experienced by the patient (for example, if a patient has one mild AE and one moderate AE, they are counted in the moderate category).

The Full Analysis Set (FAS) included all enrolled patients who attended either the Month 24 or Month 30 study visit or both in this extension study. The FAS was used for both efficacy and safety analyses.

An AE was defined as any untoward medical occurrence in a patient or clinical investigation patient administered a pharmaceutical product that did not necessarily have a causal relationship with this treatment. An AE could therefore be any unfavorable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of the medicinal (investigational) product, whether or not related to the study product.

AEs by severity (categorized as mild, moderate and severe) and relatedness (categorized as related and not related ) are considered - per each single subject - just once, at the greater severity and closest relationship to treatment. Considering that no study IMP was administered and that all standard of care were permitted, AEs must be read as related to any standard of care administered during the FU study.

The Full Analysis Set (FAS) included all enrolled patients who attended either the Month 24 or Month 30 study visit or both in this extension study. The FAS was used for both efficacy and safety analyses.

Slit-lamp biomicroscopy was performed in the study eye at Baseline and Week 8 of the NGF0120 study, and at Month 24 and 30. The anterior segment was evaluated under magnification using standard slit-lamp illumination. Parameters included eyelid edema and erythema, lashes, conjunctiva edema and bulbar erythema, cornea edema, endothelial and epithelial changes, lens, sclera, iris, and anterior chamber cells and flare. Findings were assessed visually as normal, abnormal not clinically significant, or abnormal clinically significant, based on clinical judgment and standard ophthalmologic criteria. Only the participants reporting normal and abnormal clinically significant results are shown (along with the percentage vs the "Number Analyzed", for each timepoint) The number of participants with abnormal not clinically significant results can be derived as difference between the sum of the said two categories and the "Number Analyzed", per each parameter and timepoint

The Full Analysis Set (FAS) included all enrolled patients who attended either the Month 24 or Month 30 study visit or both in this extension study. The FAS was used for both efficacy and safety analyses.