Round 1

Reviewer 1 Report (Previous Reviewer 1)

Comments and Suggestions for Authors

The author s have not resolved my main concern as indicated in the previous version. The propiate Outcome level in a study on the composition of a protein supplement is measuring the amino acid levels. Levels of indivifdual amono acids can be too high ot too low. Knowing AA levels is vital.

Measuring weight gain is not a safety outcome parameter.So, the study design was not the  correct design for this study. Therefore the study can not be accepted

Author Response

The author s have not resolved my main concern as indicated in the previous version. The propiate Outcome level in a study on the composition of a protein supplement is measuring the amino acid levels. Levels of indivifdual amono acids can be too high ot too low. Knowing AA levels is vital.

Measuring weight gain is not a safety outcome parameter.So, the study design was not the  correct design for this study. Therefore the study can not be accepted

Response: We thank the reviewer for the comments. We had provided a detailed response to the reviewer comments supported by the two references. Accordingly, the following changes were made in the revised manuscript (Discussion - limitation section):

While we assessed various pre-specified 'critical' and 'important' outcomes as directed by the COMET (Core Outcome Measures in Effectiveness Trials) initiative [51,52], we recognize that more direct metabolic endpoints, such as plasma amino acid profiles, were not evaluated primarily due to resource limitations. Plasma amino acid profiles would offer crucial insights into the potential causal pathway leading to clinically significant outcomes, such as growth. These are key points that should be focused in future RCTs.

51. Webbe JWH, Duffy JMN, Afonso E, Al-Muzaffar I, Brunton G, Greenough A, Hall NJ, Knight M, Latour JM, Lee-Davey C, et al. Core outcomes in neonatology: development of a core outcome set for neonatal research. Arch Dis Child Fetal Neonatal Ed 2020, 105, 425-431.
52. Damhuis SE, Bloomfield FH, Khalil A, Daly M, Ganzevoort W, Gordijn SJ. A Core Outcome Set and minimum reporting set for intervention studies in growth restriction in the NEwbOrN: the COSNEON study. Pediatr Res 2021, 89, 1380-1385.

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

This study seems to be already reviewed and in its current form is of interest. I have a few comments and concerns that may help further enhance the study. This is a preliminary study and the authors should address the feasibility safety and efficacy but the authors stay long in the temporary higher wight gain that may be misleading. I counsel authors with their delicate statistics to give more emphasis in the other aims of this study namely safety and feasibility.

Apparently babies with a GA< 34 weeks corrected age are fed by tube (or cup?) but not ad libitum. This is a growth study but the authors give very few information regarding volumes of food and calories intake. Where all the neonates fed with human milk during the entire period? What about special preterm infants’ formula? In how many preterm was given and at what time period and in which proportion relative to human milk? All neonates were finally fed with 180ml/kg/d really? Not individualization? Neonates with BPD for example or after sepsis need much more energy, the ESPGAN has an upper volume limit of 200ml/kg/d but reality tells that many of these sick neonates with apparent ex-utero growth restriction need much more volumes (and energy) to compensate and obtain positive balance. Furthermore, we need information regarding the growth percentiles at birth and at the end of the study and possibly at weekly interval. The higher weight gain could be beneficial if the groups were at 40 weeks ex utero growth restricted (aka the group1 became less restricted) or possibly detrimental (if the 2 groups where already well grown).How many units of fortification took each group during the entire period? the two formulations contain different kcals and as a result may have affected the final results as the plant derived formula is denser than the bovine one. 1g of HMF in 25 mL of milk  accounts to 8kcal per day or 450 kcal excess at between 32-40 weeks period which is a remarkable difference. This difference can well interpret the weight differences Head circumference is a more crucial parameter. A new table with all these details would enhance and improve the study. We again counsel authors to limit their exposure in those small weight differences and focus in the other aspects of their study

All the above reservation should be addressed by the authors before the manuscript published.

Author Response

This study seems to be already reviewed and in its current form is of interest. I have a few comments and concerns that may help further enhance the study. This is a preliminary study and the authors should address the feasibility safety and efficacy but the authors stay long in the temporary higher wight gain that may be misleading. I counsel authors with their delicate statistics to give more emphasis in the other aims of this study namely safety and feasibility.

Response: We thank the Reviewer for the valid suggestion. We have modified relevant sections in the abstract, discussion and conclusion accordingly –

Abstract - The plant-derived amino-acid based HMF demonstrated clinical feasibility and was well-tolerated. While it was associated with better in-hospital weight gain, differences in overall macronutrient density between the formulations may account for this finding. These results are hypothesis-generating and warrant future adequately powered trials.

Discussion - In this pilot RCT, we assessed the feasibility, safety, and preliminary efficacy re-garding growth parameters of a novel plant-derived amino-acid-based HMF in com-parison to the conventionally used bovine-derived whole protein HMF in preterm ne-onates born at less than 34 weeks of gestation. The results of this RCT, including the primary outcomes, are exploratory and require validation in future definitive trials. 

Conclusion - In conclusion, this pilot RCT demonstrates the feasibility and tolerability of a novel plant-derived amino-acid based HMF when compared to bovine-derived protein based HMF. The observed in-hospital weight gain benefit with plant-derived amino-acid based HMF is hypothesis-generating and may reflect its higher macronutrient density.

Apparently babies with a GA< 34 weeks corrected age are fed by tube (or cup?) but not ad libitum. This is a growth study but the authors give very few information regarding volumes of food and calories intake. Where all the neonates fed with human milk during the entire period? What about special preterm infants’ formula? In how many preterm was given and at what time period and in which proportion relative to human milk?

Response: We express our gratitude to the reviewer for the insightful comments. According to our standardized unit policy, preterm neonates in the NICU are provided with scheduled measured enteral feeding using the mother's own milk. In cases where this is unavailable, pasteurized donor human milk is utilized. Special preterm formula is not accessible in our region; therefore, standard preterm formula was employed only when the mother's own milk was insufficient post-discharge. The post-discharge period was not strictly regulated due to feasibility constraints. This aspect was addressed in the discussion section - 

“While the period from discharge until 40 weeks PMA was uncontrolled, this transient effect is consistent with that reported in literature, where post-discharge cessation of fortification due to non-compliance or other reasons such as transitioning to exclusive breast feeding typically leads to convergence in growth trajectories.”

“We could not compare the nutrient intake accurately in some neonates after discharge from the hospital as these neonates were transitioned to direct breast feeding with or without variable amounts of formula feeds.”

In response to the reviewer's queries, we have included additional elements in the methods section of the revised manuscript.

Methods - Feeds were not administered ad libitum during the period of stay in the NICU as per the unit policy. Enteral nutrition was delivered through an orogastric tube at two-hour intervals, progressing to paladai (cup) feeding when the neonate attained 31-32 weeks' PMA. The transition to direct breastfeeding commenced upon the establishment of suck-swallow-breathe coordination after 34 weeks' PMA, ensuring a balance between direct breastfeeding and sufficient nutrient intake through the use of HMF. MoM was the preferred first choice of milk and if inadequate, PDHM was used to meet the target daily intake. Standard preterm formula was used when MoM was insufficient after discharge from the NICU. During the NICU stay, the nutrient intake and type of milk were strictly controlled; however, maintaining the same was not feasible after discharge in our setting.

All neonates were finally fed with 180ml/kg/d really? Not individualization? Neonates with BPD for example or after sepsis need much more energy, the ESPGAN has an upper volume limit of 200ml/kg/d but reality tells that many of these sick neonates with apparent ex-utero growth restriction need much more volumes (and energy) to compensate and obtain positive balance.

Response: We thank the reviewer for the queries. One of the outcomes of the study was ‘time to reach enteral feed volume of 180 mL/kg/d’. We had indicated in the methods section that maximal enteral intake volume of 200 ml/kg/day was targeted (The volume of the enteral feeds was increased daily by 24 mL/kg/d in neonates of <28 weeks’ gestation and 36 mL/kg/d in others till a maximum of 200 mL/kg/d). The revised manuscript incorporates the aspect concerning the individualization of nutrient intake.

Methods - In accordance with the updated guidelines of the European Society of Pediatric Gastroenterology, Hepatology, and Nutrition, the nutrient intake for all neonates was calculated weekly to ensure the administration of maximal enteral nutrition doses in both groups.

Furthermore, we need information regarding the growth percentiles at birth and at the end of the study and possibly at weekly interval. The higher weight gain could be beneficial if the groups were at 40 weeks ex utero growth restricted (aka the group1 became less restricted) or possibly detrimental (if the 2 groups where already well grown).

Response: We concur entirely with the reviewer's perspectives. The proportion of neonates born SGA is presented in Table 1, and the proportion of neonates with EUGR at 40 weeks' PMA was one of the outcomes assessed. Both metrics were derived from a cross-sectional definition. No differences were observed in the rates of SGA at baseline and EUGR at 40 weeks' PMA between the two groups. These findings have been incorporated into the text of the revised manuscript.

Results - Initially, the proportion of neonates classified as Small for Gestational Age (SGA; weight below the 10th percentile) was 7.1% in the Bovine group compared to 12.1% in the Plant group (p=0.49). By 40 weeks postmenstrual age (PMA), the incidence of EUGR (weight below the 10th percentile) did not differ significantly between the two cohorts, with rates of 14.0% and 18.3%, respectively (p=0.63).

How many units of fortification took each group during the entire period? the two formulations contain different kcals and as a result may have affected the final results as the plant derived formula is denser than the bovine one. 1g of HMF in 25 mL of milk  accounts to 8kcal per day or 450 kcal excess at between 32-40 weeks period which is a remarkable difference. This difference can well interpret the weight differences Head circumference is a more crucial parameter. A new table with all these details would enhance and improve the study. We again counsel authors to limit their exposure in those small weight differences and focus in the other aspects of their study

All the above reservation should be addressed by the authors before the manuscript published.

Response – We express our gratitude to the reviewer for highlighting these significant concerns and concur with the perceptive critique provided. Due to feasibility constraints, we did not quantify the total number of HMF units utilized in both groups. Following the reviewer's recommendations, we have revised the discussion section of the manuscript to clarify that this was a pragmatic trial comparing two commercially available HMF formulations. We have also indicated that the observed increase in weight gain may be attributed to the superior macronutrient density of the amino acid based HMF formulation. Furthermore, we have underscored that Head Circumference, an indicator of brain growth, showed no difference, thereby supporting the reviewer's point regarding the interpretation of weight gain.

Discussion - An important interpretive consideration is that the observed difference in weight gain may be partially attributed to the higher energy density of the plant-derived HMF (4.5 kcal/g compared to 3.4 kcal/g) and its greater protein equivalent (0.35 g/g versus 0.27 g/g). With the standardized fortification strategy employed in this RCT, the plant HMF group likely received additional calories and proteins compared to the bovine HMF group. This cumulative nutrient surplus could have independently contributed to the observed weight gain advantage. Moreover, head circumference, a crucial indicator of brain growth, did not differ significantly between the groups (p=0.72), suggesting that the weight gain advantage may be due to non-neurological tissue accretion. Future research should aim to distinguish the effects of protein source from those of overall energy density. Moreover, the nutritional profile of amino-acid-based HMF, characterized by a higher content of essential fatty acids and a lower carbohydrate content, may have also contributed to improved in-hospital weight gain [42,43].

Author Response File: Author Response.pdf

Reviewer 3 Report (New Reviewer)

Comments and Suggestions for Authors

Thank you for the opportunity to review "Plant-Derived Amino-Acid vs. Bovine-Derived Protein Human Milk Fortifier in Preterm Infants < 34 Weeks’ Gestation: An Open-Label Pilot Randomized Controlled Trial".

The manuscript presents a very interesting study on the comparison of early effects of the use of two HMF with protein content from two different origins and with different biological characteristics.

The Background is very sound and clear. A minor correction: the sentence "intact bovine proteins may be poorly digested by preterm neonates due to decreased intestinal proteolytic activity" is inaccurate, because the "intestinal proteolytic activity" does not suffer a "decrease"; it should state "due to low intestinal proteolytic activity".

The Methodology is adjusted to the Aims, complete and well explained.

The Results are quite complete and well understood in the text but their presentation in tables must be improved (explained in notes to the manuscript). Specifically, Table 2. should rather present the description of the data for each parameter and group, then the MD and after the aMD (both with the comparisons p-value). A minor correction: the sentence "mortality and other morbidities were also comparable between the two groups" is inaccurate, because those outcomes are indeed comparable, in fact, the Authors compare them and conclude they are "similar" or "not significantly different", and should be referred as such.

The Discussion is well written, adequately humble, but it misses references to the absent option of HMF with protein based on whey protein hydrolysate and caseinate hydrolysate from cow’s milk (as Nutrilon, Nutricia Human Milk Fortifier). Eventually those are not available in the country but the Authors may consider to include a reference.

The overall impression of the version of the manuscript I read is very positive.

Comments for author File: Comments.pdf

Author Response

Thank you for the opportunity to review "Plant-Derived Amino-Acid vs. Bovine-Derived Protein Human Milk Fortifier in Preterm Infants < 34 Weeks’ Gestation: An Open-Label Pilot Randomized Controlled Trial".

The manuscript presents a very interesting study on the comparison of early effects of the use of two HMF with protein content from two different origins and with different biological characteristics.

The Background is very sound and clear. A minor correction: the sentence "intact bovine proteins may be poorly digested by preterm neonates due to decreased intestinal proteolytic activity" is inaccurate, because the "intestinal proteolytic activity" does not suffer a "decrease"; it should state "due to low intestinal proteolytic activity".

Response: We thank the reviewer for the encouraging remarks. We have corrected the word in the revised manuscript.

The Methodology is adjusted to the Aims, complete and well explained.

The Results are quite complete and well understood in the text but their presentation in tables must be improved (explained in notes to the manuscript). Specifically, Table 2. should rather present the description of the data for each parameter and group, then the MD and after the aMD (both with the comparisons p-value).

Response – We appreciate the reviewer's comments and concur with the reviewer’s perspective. Consequently, we have modified the structure of Table 2 in the revised manuscript:

A minor correction: the sentence "mortality and other morbidities were also comparable between the two groups" is inaccurate, because those outcomes are indeed comparable, in fact, the Authors compare them and conclude they are "similar" or "not significantly different", and should be referred as such.

Response – We thank the reviewer for pointing this. We have revised the sentence accordingly.

The Discussion is well written, adequately humble, but it misses references to the absent option of HMF with protein based on whey protein hydrolysate and caseinate hydrolysate from cow’s milk (as Nutrilon, Nutricia Human Milk Fortifier). Eventually those are not available in the country but the Authors may consider to include a reference.

Response: We appreciate the reviewer's positive feedback. Additionally, we have incorporated the potential application of hydrolyzed protein-based HMF in the discussion section.

Discussion - In this context, plant-derived amino-acid HMFs may offer a more scalable and practical alternative for use in LMICs. Additionally, alternative fortifiers, such as those based on extensively hydrolyzed bovine protein, have been assessed [23]. Although these ex-tensively hydrolyzed bovine protein HMFs exhibit reduced antigenicity compared to intact whey, they are not available in our region.

The overall impression of the version of the manuscript I read is very positive.

Response – We are thankful to the reviewer for the positive feedback.

Author Response File: Author Response.pdf

Reviewer 4 Report (New Reviewer)

Comments and Suggestions for Authors

This is a markedly strong and well-constructed manuscript. It is an open-label, two-center pilot RCT comparing a plant-derived amino-acid HMF against a standard bovine whey-protein HMF in preterm infants <34 weeks. The trial is prospectively registered, follows CONSORT-for-pilot guidance, uses a pre-specified analysis plan with sensible missing-data handling (MICE/PMM, Rubin’s rules), and is appropriately framed as hypothesis-generating. The author’s conclusions are appropriate.

Issues to address:

1. The abstract’s CI does not match the body. The abstract reports the primary adjusted weight-gain result as “aMD 3.20 g/Kg/day, 95% CI 0.46–5.95.” But the subgroup/Results sup table 1 section reports the overall adjusted ITT estimate as “3.20 g/Kg/d (0.48 to 5.92); p=0.02,” and Supplement Table 1 gives “3.20 (0.48, 5.92).” The same overall estimate appears with two different confidence intervals (0.46–5.95 vs. 0.48–5.92) in different places. These must be reconciled.
2. Multiplicity is a real concern and is under-discussed. There are two co-primary growth time points (discharge, 40w PMA) across three anthropometric measures, plus a time-to-feed primary, plus numerous secondary outcomes, plus post-hoc subgroup analyses by GA and BW. The single positive finding emerges and no multiplicity adjustment is mentioned and should be.
3. This compares two whole formulations with different macronutrients. The authors acknowledge this, but it is arguably the single most important interpretive limitation and deserves more prominence. The plant HMF delivers more energy (4.5 vs. 3.4 kcal/g), more protein equivalent (0.35 vs. 0.27 g/g), more fat, added DHA/ARA, and lower osmolality. Any weight-gain difference cannot be attributed to “plant-derived amino acids” per se; it may simply reflect higher protein/energy density. The title and abstract should reflect this.
4. Discrepancy in the enrollment dates. Methods state enrollment ran “from 19th June 2025 to 1st February 2026.” The Results state “assessed for eligibility from 19th June 2025 to 1st January 2026.” 
5. Secondary outcomes table presents non-integer event counts. Table 3 reports categorical counts such as “RBC transfusion 11.4 (16.2%),” “LONS 22.1 (31.6%),” “BPD 5.1 (7.3%).” Fractional counts presumably reflect pooling across 50 imputed datasets, but presenting imputed fractional counts for categorical safety outcomes is unconventional.  Readers expect observed counts for events like NEC and death. Consider reporting observed event counts for safety outcomes (imputation used only for effect estimates) and explain the fractional values explicitly.
6. Safety outcomes are underpowered and should not be described as reassuring without caveat. NEC events are tiny (2 vs. 3) and mortality similarly (2 vs. 3), with large CIs. “Comparable safety profile” is technically accurate) but the trial cannot exclude meaningful harm. Since the entire rationale for moving away from bovine HMF is NEC/mortality reduction, the inability to say anything about the outcome that matters most should be stated plainly; the abstract’s “comparable safety profile” overstates.
7. The quality of weight gain (lean vs. fat mass) is not addressed in the limitations. The trial’s headline benefit is a higher in-hospital weight-gain velocity, but weight gain is reported only as total mass; its composition — accretion of lean (fat-free) mass versus fat mass — is neither measured nor discussed. The two fortifiers differ in macronutrients: the plant amino-acid product is higher in protein equivalent and fat and lower in carbohydrate than the bovine additive. Higher protein intake tends to favor lean-mass accretion, whereas excess energy from fat/carbohydrate tends to favor fat-mass accretion, so an identical total weight gain could reflect quite different body-composition trajectories between arms. This matters clinically: infant body composition (not weight alone) is the more accurate marker of nutritional quality and is associated with later cardiometabolic and neurodevelopmental outcomes, with feeding type and protein intake among the recognized determinants (Jerome et al., Nutr Clin Pract 2023, PMID: 37721459). At minimum, the Discussion should state that the weight-gain advantage cannot be interpreted as beneficial without knowing whether it represents lean or fat mass.
8. Numerous portions of text are red. I am assuming this is an artifact of editing and they should not appear this way in the final text.
9. Affiliation numbering error. Five numbered affiliations are listed, but author superscripts use only 1–4. Affiliation 5 (Owaisi Hospital/Deccan College) is not linked to any author.

Grammatical and Language Issues

The writing is generally clear and professional. Issues are minor and mostly typographical:

1. “velocitiy” appears twice in the sentence reporting the primary result (“weight gain velocitiy from enrolment until discharge”) — misspelling of “velocity.”
2. “anlyses” / “Undjusted” / “anthrpometric” in Supplement Figures 4–6 headers (e.g., “ITT- Adjusted anlyses”; “Sensitivity Analyses of anthrpometric variables (PP − Undjusted analyses)”). Multiple spelling errors embedded in figure graphics needs correcting.
3. Inconsistent unit notation. “g/Kg/day,” “g/Kg/d,” “gm/Kg/d,” and “g/kg/d” all appear; capitalized “Kg” is nonstandard (SI is lowercase “kg”). Standardize to “g/kg/d” and “mL/kg/d.”
4. “Bells’ staging” — should be “Bell’s staging” 
5. “weight for age of less than 10th centile” — should read “less than the 10th centile.”
6. CI punctuation. The manuscript mixes dash, comma, and “to” separators for confidence intervals (“0.46–5.95,” “(0.48, 5.92),” “(0.48 to 5.92)”). 

Author Response

This is a markedly strong and well-constructed manuscript. It is an open-label, two-center pilot RCT comparing a plant-derived amino-acid HMF against a standard bovine whey-protein HMF in preterm infants <34 weeks. The trial is prospectively registered, follows CONSORT-for-pilot guidance, uses a pre-specified analysis plan with sensible missing-data handling (MICE/PMM, Rubin’s rules), and is appropriately framed as hypothesis-generating. The author’s conclusions are appropriate.

Response – We thank the reviewer for the comments.

Issues to address:

1. The abstract’s CI does not match the body. The abstract reports the primary adjusted weight-gain result as “aMD 3.20 g/Kg/day, 95% CI 0.46–5.95.” But the subgroup/Results sup table 1 section reports the overall adjusted ITT estimate as “3.20 g/Kg/d (0.48 to 5.92); p=0.02,” and Supplement Table 1 gives “3.20 (0.48, 5.92).” The same overall estimate appears with two different confidence intervals (0.46–5.95 vs. 0.48–5.92) in different places. These must be reconciled.

Response – We apologize for this clerical error. The exact point estimate with 95% CI from the pooled multiple imputation analysis is 3.20 g/kg/d (95% CI 0.46 to 5.95). We have uniformly standardized this throughout the Abstract, Results, and Supplementary Tables.

1. Multiplicity is a real concern and is under-discussed. There are two co-primary growth time points (discharge, 40w PMA) across three anthropometric measures, plus a time-to-feed primary, plus numerous secondary outcomes, plus post-hoc subgroup analyses by GA and BW. The single positive finding emerges and no multiplicity adjustment is mentioned and should be.

Response – We agree with the reviewer’s point. In sync with that we have highlighted in relevant sections of the manuscript regarding the pilot nature of this RCT and that the findings are exploratory. The following have been included in the revised manuscript –

Abstract - The plant-derived amino-acid based HMF demonstrated clinical feasibility and was well-tolerated. While it was associated with better in-hospital weight gain, differences in overall macronutrient density between the formulations may account for this finding. These results are hypothesis-generating and warrant future adequately powered trials.

Discussion - In this pilot RCT, we assessed the feasibility, safety, and preliminary efficacy regarding growth parameters of a novel plant-derived amino-acid-based HMF in com-parison to the conventionally used bovine-derived whole protein HMF in preterm neonates born at less than 34 weeks of gestation. The results of this RCT, including the primary outcomes, are exploratory and require validation in future definitive trials. 

Conclusion - The observed in-hospital weight gain benefit with plant-derived amino-acid based HMF is hypothesis-generating and may reflect its higher macronutrient density. These pre-liminary findings inform the design of future multi-center RCTs adequately powered to assess critical safety outcomes.

1. This compares two whole formulations with different macronutrients. The authors acknowledge this, but it is arguably the single most important interpretive limitation and deserves more prominence. The plant HMF delivers more energy (4.5 vs. 3.4 kcal/g), more protein equivalent (0.35 vs. 0.27 g/g), more fat, added DHA/ARA, and lower osmolality. Any weight-gain difference cannot be attributed to “plant-derived amino acids” per se; it may simply reflect higher protein/energy density. The title and abstract should reflect this.

Response: We agree with reviewer’s view. We have further elaborated on this aspect in the abstract and discussion sections of the revised manuscript.

Abstract - While it was associated with better in-hospital weight gain, differences in overall macronutrient density between the formulations may account for this finding.

Discussion - It is crucial to emphasize that this trial compared two distinct formulations, not just protein sources. An important interpretive consideration is that the observed difference in weight gain may be partially attributed to the higher energy density of the plant-derived HMF (4.5 kcal/g compared to 3.4 kcal/g) and its greater protein equivalent (0.35 g/g versus 0.27 g/g). With the standardized fortification strategy employed in this RCT, the plant HMF group likely received additional calories and proteins compared to the bovine HMF group. This cumulative nutrient surplus could have independently contributed to the observed weight gain advantage. Moreover, head circumference, a crucial indicator of brain growth, did not differ significantly between the groups (p=0.72), suggesting that the weight gain advantage may be due to non-neurological tissue accretion. Future research should aim to distinguish the effects of protein source from those of overall energy density. Moreover, the nutritional profile of amino-acid-based HMF, characterized by a higher content of essential fatty acids and a lower carbohydrate content, may have also con-tributed to improved in-hospital weight gain.

1. Discrepancy in the enrollment dates. Methods state enrollment ran “from 19th June 2025 to 1st February 2026.” The Results state “assessed for eligibility from 19th June 2025 to 1st January 2026.” 

Response: We appreciate the reviewer's inquiry. To clarify, although enrolment concluded on 1 January 2026, the follow-up for all enrolled neonates was completed by 1 February 2026.

1. Secondary outcomes table presents non-integer event counts. Table 3 reports categorical counts such as “RBC transfusion 11.4 (16.2%),” “LONS 22.1 (31.6%),” “BPD 5.1 (7.3%).” Fractional counts presumably reflect pooling across 50 imputed datasets, but presenting imputed fractional counts for categorical safety outcomes is unconventional.  Readers expect observed counts for events like NEC and death. Consider reporting observed event counts for safety outcomes (imputation used only for effect estimates) and explain the fractional values explicitly.

Response: We thank the Reviewer for the guidance; fractional counts were an artifact of pooling across 50 imputed datasets via Rubin's rules. We have updated Table 3 to present the transparent, raw observed integer event counts for the cohorts, using the imputation models strictly for generating the final Adjusted Risk Ratios (aRR) and Confidence Intervals.

1. Safety outcomes are underpowered and should not be described as reassuring without caveat. NEC events are tiny (2 vs. 3) and mortality similarly (2 vs. 3), with large CIs. “Comparable safety profile” is technically accurate) but the trial cannot exclude meaningful harm. Since the entire rationale for moving away from bovine HMF is NEC/mortality reduction, the inability to say anything about the outcome that matters most should be stated plainly; the abstract’s “comparable safety profile” overstates.

Response: We thank the reviewer’s critical point. We have indicated in the previous response (reviewer query 2) that we have indicated the pilot nature of this RCT and that the effect estimates including that for growth outcomes are not confirmatory and the need for definitive trials. We have revised the manuscript to further emphasize these. We have also removed terms like "comparable safety profile".

Results - The other secondary outcomes of days to regain birth weight, number of feed intolerance days, duration of hospital stay, and safety outcomes of mortality and other morbidities were also not significantly different between the two groups.

Discussion (limitation) - As a pilot study, the sample size was notably inadequate to rule out significant harm or to identify differences in rare but critical safety outcomes, such as NEC and mortality. Consequently, the safety data presented here are preliminary and do not indicate safety equivalence.

1. The quality of weight gain (lean vs. fat mass) is not addressed in the limitations. The trial’s headline benefit is a higher in-hospital weight-gain velocity, but weight gain is reported only as total mass; its composition — accretion of lean (fat-free) mass versus fat mass — is neither measured nor discussed. The two fortifiers differ in macronutrients: the plant amino-acid product is higher in protein equivalent and fat and lower in carbohydrate than the bovine additive. Higher protein intake tends to favor lean-mass accretion, whereas excess energy from fat/carbohydrate tends to favor fat-mass accretion, so an identical total weight gain could reflect quite different body-composition trajectories between arms. This matters clinically: infant body composition (not weight alone) is the more accurate marker of nutritional quality and is associated with later cardiometabolic and neurodevelopmental outcomes, with feeding type and protein intake among the recognized determinants (Jerome et al., Nutr Clin Pract 2023, PMID: 37721459). At minimum, the Discussion should state that the weight-gain advantage cannot be interpreted as beneficial without knowing whether it represents lean or fat mass.

Response: We thank the reviewer for the suggestions and providing the relevant reference to validate the same. We agree completely with the reviewer’s views. Accordingly we have modified the manuscript and incorporated the relevant reference.

Discussion (limitation) - A notable limitation of this study is the assessment of weight gain solely in terms of total mass, without analyzing the specific composition of this gain, particularly the distinction between lean (fat-free) mass and fat mass. Infant body composition which is significantly influenced by variations in protein and energy intake is a more precise indicator of nutritional quality and a superior predictor of long-term neurodevelopmental and cardiometabolic outcomes compared to weight alone [53].

Reference 53 - Jerome ML, Valcarce V, Lach L, Itriago E, Salas AA. Infant body composition: A comprehensive overview of assessment techniques, nutrition factors, and health outcomes. Nutr Clin Pract. 2023, 38 Suppl 2(Suppl 2), S7-S27.

1. Numerous portions of text are red. I am assuming this is an artifact of editing and they should not appear this way in the final text.

Response: In accordance with the journal's requirements, we submitted a marked version of the revised manuscript, with revisions highlighted in red font. The current version of the manuscript also contains marked sections that specifically indicate the changes made in response to the reviewer's suggestions for this round of review.

1. Affiliation numbering error. Five numbered affiliations are listed, but author superscripts use only 1–4. Affiliation 5 (Owaisi Hospital/Deccan College) is not linked to any author.

Response: We express our gratitude to the reviewer for highlighting this issue. It has been addressed in the revised manuscript.

Grammatical and Language Issues

The writing is generally clear and professional. Issues are minor and mostly typographical:

1. “velocitiy” appears twice in the sentence reporting the primary result (“weight gain velocitiy from enrolment until discharge”) — misspelling of “velocity.”

1. “anlyses” / “Undjusted” / “anthrpometric” in Supplement Figures 4–6 headers (e.g., “ITT- Adjusted anlyses”; “Sensitivity Analyses of anthrpometric variables (PP − Undjusted analyses)”). Multiple spelling errors embedded in figure graphics needs correcting.

1. Inconsistent unit notation. “g/Kg/day,” “g/Kg/d,” “gm/Kg/d,” and “g/kg/d” all appear; capitalized “Kg” is nonstandard (SI is lowercase “kg”). Standardize to “g/kg/d” and “mL/kg/d.”

1. “Bells’ staging” — should be “Bell’s staging” 

1. “weight for age of less than 10th centile” — should read “less than the 10th centile.”

1. CI punctuation. The manuscript mixes dash, comma, and “to” separators for confidence intervals (“0.46–5.95,” “(0.48, 5.92),” “(0.48 to 5.92)”). 

Response: We express our gratitude to the reviewer for the meticulous examination of our manuscript's formatting. All typographic errors have been rectified, and we have consistently used 'g/kg/d' and 'mL/kg/d' throughout the manuscript. Additionally, all 95% confidence intervals are now indicated with commas.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report (Previous Reviewer 1)

Comments and Suggestions for Authors

In the paper. In the response the authors refer to papers, but they did not respond to my concern..
this paper is only acceptable when the authors delete any reference that the present study showed that the plant derived for tyfyer is safe. The whole claim about safe must be deleted.

Author Response

In the paper. In the response the authors refer to papers, but they did not respond to my concern..
this paper is only acceptable when the authors delete any reference that the present study showed that the plant derived for tyfyer is safe. The whole claim about safe must be deleted.

Response: We thank the reviewer for the comments. We had adequately addressed the concerns of the reviewer in the revised draft. 

Abstract-

Conclusions: The plant-derived amino-acid based HMF demonstrated clinical feasibility and was well-tolerated. While it was associated with better in-hospital weight gain, differences in overall macronutrient density between the formulations may account for this finding. These results are hypothesis-generating and warrant future adequately powered trials.

Discussion- 

Plasma amino acid profiles would offer crucial insights into the potential causal pathway leading to clinically significant outcomes, such as growth. These are key points that should be focused in future RCTs.

Finally, as a pilot study, the sample size was notably inadequate to rule out significant harm or to identify differences in rare but critical safety outcomes, such as NEC and mortality. Consequently, the safety data presented here are preliminary and do not indicate safety equivalence. 

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

the authors has replied satisfactorily in all concerns and suggestions 

Author Response

We thank the reviewer for the valuable suggestions and the encouraging response.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is a study on growth parameters of preterm infants receiving a human milk supplement based on bovine sources or vegetable sources. The conclusion is that there is no difference in growth parameters between both groups. This is reassuring and interesting. Studies on the optimal human milk fortifyers are needed. I do have the following comments.

1. The AA composition of both the vegetable and bovine sources is not given.
2. The outcome parameters in this study , growth, are not the most important outcome parameters. This study looks at growth, but the real important outcome is the aa compostion in plasma. 
3. this study is not designed to study safety, growth is not a safety measure.
4. It  is essential to show the AA profile in both groups and in a group fed human milk without a fortifyer

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you very much for the invitation to review the manuscript entitled “Plant-Derived Amino-Acid vs. Bovine-Derived Protein Human Milk Fortifier in Preterm Infants <34 Weeks’ Gestation: A Pilot Randomized Controlled Trial.” I appreciate the opportunity to evaluate this timely and clinically relevant study. Below, I provide my detailed assessment. This manuscript reports an open-label, parallel-group pilot randomized controlled trial comparing a plant-derived amino-acid–based human milk fortifier (HMF) with a bovine-derived whole protein HMF in preterm infants <34 weeks’ gestation. The primary outcomes included time to reach full enteral feeds (180 mL/kg/day) and growth velocities to discharge and 40 weeks’ PMA, with additional safety and feasibility endpoints. The topic is highly relevant. Optimization of protein supplementation in preterm infants remains a major clinical priority, particularly in low- and middle-income countries where access to human milk–derived fortifiers is limited. The evaluation of an amino-acid–based plant-derived HMF in a randomized framework represents a meaningful and novel contribution. The study is generally well conducted and clearly written. However, several methodological, statistical, and interpretative issues require clarification or refinement before the manuscript can be considered for publication.

Comment 1: The manuscript appropriately identifies the study as a pilot RCT; however, the interpretation of efficacy outcomes is somewhat overstated. While an adjusted mean difference of +3.20 g/kg/day in in-hospital weight gain is statistically significant (adjusted analysis), the study was not powered for efficacy outcomes. The authors should explicitly state that the trial was not powered to detect clinically meaningful differences in growth or morbidity; the conclusion that the plant-derived HMF “is a potential alternative” should be more cautiously framed as hypothesis-generating; a clearer distinction between feasibility outcomes and exploratory efficacy findings is needed in both the Results and Discussion.

Comment 2: Given the open-label design, clinicians were aware of group allocation. This may have influenced: advancement of feeds, clinical diagnosis of feed intolerance, and decision-making regarding discharge timing. Although time to full feeds did not differ, subjective endpoints (e.g., feed intolerance, duration of hospital stay) are vulnerable to performance and detection bias. The authors should: explicitly discuss potential biases arising from lack of blinding; clarify whether outcome assessors (e.g., anthropometric measurements) were blinded; describe whether standardized measurement protocols were used for weight, length, and head circumference.

Comment 3: Table 1 shows a higher proportion of umbilical artery A/REDF in the plant HMF group (13.6% vs. 4.3%). Although adjusted analyses included this covariate, the clinical implications deserve discussion, as such infants may have impaired postnatal growth trajectories. The authors should: provide p-values for baseline comparisons (currently absent); discuss the potential impact of this imbalance; consider presenting unadjusted and adjusted models side by side in the main text for transparency.

Comment 4: The use of MICE with 50 imputations is commendable. However: the proportion of missing data per variable is not reported; it is unclear whether missingness was assumed to be Missing at Random (MAR), and whether this assumption is plausible; sensitivity analyses comparing complete-case vs. imputed results would strengthen credibility. Please provide: a table summarizing missing data by group and variable; a brief justification of the MAR assumption, a short sensitivity comparison in supplementary material.

Comment 5: Exploratory subgroup analyses (e.g., BW ≥1500 g) are highlighted in the abstract and discussion. However: the study is clearly underpowered for interaction testing, no formal interaction p-values are presented; the risk of spurious findings is high. Subgroup findings should be clearly labelled as exploratory and interpreted with strong caution.

Comment 6: The adjusted increase of 3.20 g/kg/day in weight gain from enrolment to discharge is statistically significant. However: what was the absolute duration of exposure? What is the estimated cumulative weight difference at discharge? Does this translate into clinically meaningful improvements in z-scores or EUGR rates? Given that EUGR at 40 weeks was similar, the authors should contextualize whether the observed difference is clinically meaningful or transient.

Comment 7: The two HMF products differ not only in protein source but also in macronutrient composition (energy, carbohydrate, fat, micronutrients). This complicates attribution of effects to: amino acid composition; osmolality, total energy density; protein dose. The manuscript should explicitly acknowledge that this is not an isolated protein-source comparison but a comparison of two distinct formulations.

Comment 8: Consider adding “Open-Label” in the title to improve transparency.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The aim of this study is to compare two human milk fortifyers, one from animal origin and one from vegetable origin. My main concern is that the investigators in my opinion choose wrong outcome parameters. Now they used weight gain and the prevalence of diseases NEC and BPD. In my opinion the correct outcome data are AA levels in plasma. Plasma levels either can be too low or too high. Both are dangerous. I do not feel that incidence of diseases is a good outcome measure for e nutritional intervention. And, parameters like weight gain are also dependent on the intake of calories, fatty acids and minerals. Data on these compounds are missing in this study. A better design of the study would have been to compare plasma levels in the two groups, bovine supplement and vegetable supplement. Now only a list of AA in the vegetable group is given, but no information on the amount provided in each group. 

The idea of this study is good, but the study design is not  sufficient to make conclusions. 

Author Response

We thank the Reviewer for the valuable comments.

1. We acknowledge the reviewer's perspective concerning the study's findings and the emphasis on the specific outcome of 'plasma amino acid levels.' We concur that plasma amino acid profiles could offer valuable mechanistic insights. However, for a pilot trial centred on feasibility and clinically meaningful outcomes, we prioritized outcomes that are of significant importance to patients and are recommended by COMET (Core Outcome Measures in Effectiveness Trials) Initiative. Measuring plasma amino acids was beyond the scope of this resource-limited study, but we have included this as a recommendation for future research.. Our comprehensive response is as follows:

• The outcome 'plasma amino acid levels' was considered either a 'surrogate outcome' or a 'not important outcome.' This perspective is informed by studies conducted by the COMET initiative, a highly regarded organization that has developed 'core outcome sets (COS)' and 'minimum reporting sets (MRS)' to guide the reporting of clinical trials across various specialties, including neonatology. Although it is not obligatory to report only the outcomes specified by the COMET group, these outcomes are regarded as the 'minimum set' that ensures neonatal research studies report a common subset of clinically significant outcomes. This standardization facilitates future meta-analyses of similar trials by ensuring uniformity in outcome reporting. While the COMET consensus on the 'MRS for measures of nutritional intake and growth in preterm studies: A Delphi study' (https://www.comet-initiative.org/Studies/Details/3185) has been completed, it has not yet been published. Nonetheless, we refer to two published papers from the COMET group that delineate the core outcomes to be considered in neonatal trials. All outcomes reported in our trial are included among the 'critical' and 'important' outcomes specified in these papers. Please find the outcomes reported in our trial (highlighted in bold) alongside the COMET consensus.
• Core outcomes in neonatology: development of a core outcome set for neonatal research (Webbe JWH, Duffy JMN, Afonso E, et al. Arch Dis Child Fetal Neonatal Ed 2020;105:F425–F431. doi:10.1136/archdischild-2019-317501) – Survival, Sepsis, Necrotising enterocolitis, Brain injury on imaging, Retinopathy of prematurity (preterm only), General gross motor ability, General cognitive ability, Quality of life, Adverse events, Visual impairment or blindness, Hearing impairment or deafness, Chronic lung disease/bronchopulmonary dysplasia (preterm only).
• A Core Outcome Set and minimum reporting set for intervention studies in growth restriction in the NEwbOrN: the COSNEON study (Damhuis SE et al., Pediatr Res. 2021;89(6):1380-1385. doi: 10.1038/s41390-020-01119-5) - Gastrointestinal (Domain): Necrotizing enterocolitis, Need for gastrointestinal surgery; Growth/weight: Head circumference, Length, Weight gain.
• We have incorporated the consideration of ‘plasma amino acid levels’ into our revised manuscript in response to the Reviewer's suggestion and as a potential direction for future research - Discussion, page 16:“While we assessed various pre-specified 'critical' and 'important' outcomes as directed by the COMET (Core Outcome Measures in Effectiveness Trials) initiative [50,51], we recognize that more direct metabolic endpoints, such as plasma amino acid profiles, were not evaluated primarily due to resource limitations. Plasma amino acid profiles would offer crucial insights into the potential causal pathway leading to clinically significant outcomes, such as growth. These are key points that should be focused in future RCTs.”

2. We concur with the Reviewer's observation that weight gain is contingent upon the intake of calories, fatty acids, and minerals. In the revised manuscript, we have included the energy content and composition of the two HMFs utilized in our study. We have thoroughly addressed the issue raised by the reviewer in the subsequent sections of the manuscript.

• Discussion, page 15: “It is crucial to emphasize that this trial compared two distinct formulations, not just protein sources. As shown in previous studies, the amino-acid based HMF's nutritional profile which is higher in protein and essential fatty acids but lower in carbohydrates may also have contributed to better in-hospital weight gain [42,43]. Furthermore, an in vitro study evaluating various HMFs, including those used in our trial, revealed through direct comparison that the amino-acid based fortifier yields a lower reconstituted osmolality compared to the bovine-based protein HMF [17]. This is probably significant as a higher osmolar load has been shown to be associated with altered intestinal mucosal integrity in pre-clinical studies which could adversely affect the nutrient absorption [19,44,45]. We acknowledge that the higher protein content in the plant-derived amino-acid based HMF in conjunction with the use of standardized fortification approach would have possibly contributed to the observed better in-hospital weight gain.”
• Discussion, page 16: “We could not compare the nutrient intake accurately in some neonates after discharge from the hospital as these neonates were transitioned to direct breast feeding with or without variable amounts of formula feeds.”

Reviewer 2 Report

Comments and Suggestions for Authors

I have carefully examined the authors’ responses and the updated manuscript. Overall, I am satisfied that the authors have addressed the major concerns raised in the previous round of review in a thorough and constructive manner. I commend the authors for their careful and thoughtful revisions.

Author Response

We express our sincere gratitude to the Reviewer for the valuable suggestions and appreciate the encouraging feedback provided.